INFORMATION TO USERS

This dissertation was produced fro m a m ic ro film copy o f the original docum ent.
W hile th e most advanced technological means to photograph and reproduce this
d ocu m en t have been used, th e q u a lity is heavily d ependent upon th e q u ality of
the original su bm itted.
The

fo llo w in g

exp lan ation

o f techniques is provided

to

help

you

understand

markings or p atterns which m ay appear on this reproduction.
1.

T h e sign o r “ ta rg e t'' fo r pages apparently lacking fro m th e docum ent
photographed

is “ Missing Page(s)". If it was possible to obtain the

missing page(s) o r section, they are spliced into th e film along w ith
adjacent pages. T h is may have necessitated cu tting th ru an image and
dup licatin g adjacent pages to insure you co m p lete c o n tin u ity .
2.

W hen an image on the film
m ark,
co py

is o bliterated w ith a large round black

it is an in d ic a tio n th a t th e photographer suspected th a t th e
may

have moved

during exposure and thus cause a blurred

image. Y o u w ill fin d a good image of th e page in th e adjacent fram e.
3.

W hen a map, drawing o r chart, etc., was p art o f th e m aterial being
p h o to g ra p h e d

th e

photographer

"section in g " the m aterial.

fo llo w e d

a

d e fin ite

m ethod

in

It is custom ary to begin photoing at the

upper le ft hand corner o f a large sheet and to co ntin u e photoing fro m
le ft

to

right

sectioning

in

equal sections w ith

a small overlap.

If necessary,

is continued again — beginning below th e firs t row and

co n tin u in g on u n til com plete.
4.

T h e m a jo rity of users indicate th a t the te x tu a l c o n ten t is of greatest
value,

however,

m ade fro m
dissertation.

a som ew hat higher q u a lity

"p ho to g rap h s"
Silver

prints

if essential
of

re p ro d uctio n

could be

to th e understanding o f the

"p ho to g rap h s"

m ay

be

ordered

at

ad dition al charge by w ritin g th e O rder D e p a rtm e n t, giving th e catalog
num ber, title , a u th o r and specific pages yo u wish reproduced.

University Microfilms
300 North Zeeb Road
Ann Arbor, M ichigan 48106
A Xerox Education Company

73-5369
FARICY, William Henry, 1930A CLASSIFICATION OF ACADB1IC DEPARTMENTS AT
MICHIGAN STATE UNIVERSITY BASED ON FUNCTIONAL
CHARACTERISTICS.
Michigan State University, Ph.D., 1972
Education, higher

University Microfilms, A XEROX Company, Ann Arbor, Michigan

(^Copyright by
WILLIAM HENRY FARICY
1972

A CLASSIFICATION OF ACADEMIC DEPARTMENTS AT
MICHIGAN STATE UNIVERSITY BASED ON
FUNCTIONAL CHARACTERISTICS

By
Will i a m Henry Faricy

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Higher Education
1972

PLEASE NOTE:

Some pages may have
indistinct p rin t.
Filmed as r e c e i v e d .

U n i v e r s i t y M i c r o f i l m s , A Xerox Education Company

f<

I

$■'

fc

I

ABSTRACT
A CLASSIFICATION OF ACADEMIC DEPARTMENTS AT
MICHIGAN STATE UNIVERSITY BASED ON
FUNCTIONAL CHARACTERISTICS
By
William Henry Faricy

Problem
University departments can be considered in terms
of six essential aspects:
governance, members,

subject matter,

resources,

functions, and objectives.

However,

departments have usually been defined in terms of academic
disciplines or subject matters and reflect current assump­
tions about the proper articulations of the body of k nowl­
edge.

The structures in which departments are organized

(colleges, divisions,

schools)

also have usually been

defined in terms of subject matters.
is only one aspect of a department,

Since subject matter
and since conceptions

of subject matter are merely assumptions,
units based on subject matter
superstructures)

the university

(both departments and their

cannot reflect the full complexity of

university operations and are essentially arbitrary.
Further, grouping departments by subject matter obscures
both the similarities and dissimilarities in departmental
functions, and inhibits valid comparisons among departments,

L

William Henry Faricy

thereby ultimately hindering the evaluation and improvement
of departmental operations.

Objectives
This study sought to analyze objective traits
describing d e p a r t m e n t s ' functions in order to determine
dimensions in the traits, and then to group the departments
on the basis of those dimensions, by means of a cluster
analysis.

These data and methods are appropriate for a

conception of the department as an organization with several
essential aspects,

and a conception of the university as an

organization that is essentially a unique entity in a
specific context.

Procedures
The population of interest consisted of 79 of the
academic departments at Michigan State University.

The data

included 167 departmental traits chiefly drawn from reports
routinely prepared by the university Office of Institutional
Research and other offices.
Cluster analysis resembles factor analysis,

but

differs from it because clusters of variables are hypoth­
esized prior to the analysis, usually on the basis of the
variables'
two parts:

correlations.

Thus, the cluster analysis had

correlation and clustering of variables, and a

multiple-groups analysis to test whether the clusters were
coherent.

The procedures involved several repetitions in

William Henry Faricy
order to refine the clusters.

The multiple-groups analysis

produced loadings that indicated the degree to which each
trait was correlated with its trait-cluster and each
department with its department group.

Results
Sixteen trait-clusters were produced on the basis
of the traits'

correlations across the 79 departments.

Then, on the basis of the departments'

correlations across

the 16 trait-clusters, eleven department groups were pro­
duced.

These groups were revised by combining groups with

appropriately high correlations and by eliminating depart­
ments with low loadings.

The final result was a set of six

groups representing departmental types-.
Science,

(b) Health Science,

(c) Service Technology,

Humanities and General Studies,
and

(d)

(e) Scientific Disciplines,

(f) Applied Science and Technology.

main classes, distinguished chiefly by
mission and

(a) Agricultural

They fall into two
(1) instructional

(2) staff utilization.

Conclusions
The procedures appeared useful for clustering both
traits and departments.

However,

the specific results,

although interesting and illustrative, were not completely
satisfactory.

Grouping the departments on the basis of the

16 trait-clusters rather than the full set of 167 traits
proved to be an error,

since individual traits were thus

William Henry Faricy

embedded within their clusters and could not be analyzed in
their individual relations to the department groups.
because of the heterogeneity of departmental traits,

Also,
it

might be more useful and meaningful to group departments
on the basis of sub-sets of traits in further studies of
this topic.

ACKNOWLEDGMENTS

Dr. Paul L. Dressel has been Invaluably helpful and
encouraging as the major advisor for this dissertation.

The

candidate has deeply appreciated the opportunity to work
with Dr. Dressel,

as assistant director of the Department

Study Project for the past two years.
Dr. Andrew C. Porter has generously offered advice
and criticism that have helped to make this dissertation a
better one.

Dr. Max R. Raines has also helped the candidate

in several w a y s .
The staff of the Office of Institutional Research
has been a constant source of support and assistance.
Sheila Hoeve has displayed outstanding competence and
patience in collecting data and typing manuscript drafts.
The Esso Educational Foundation has contributed to
this dissertation through its support of the Department
Study Project.
William J. Browne, Jr., has been a great help in
handling the computer programs.

iii

TABLE OF CONTENTS
Page
LIST OF T A B L E S ............................................

vi

LIST OF F I G U R E S ..........................................

vii

Chapter
I.

II.

III.

IV.

V.

B A C K G R O U N D ........................................

1

S u m m a r y .....................................

20

THE P R O B L E M .....................................

22

S u m m a r y ..................... . .............

30

OBJECTIVES, DATA, PROCEDURES

...................

32

O b j e c t i v e s ...................................
Data:
Population and T r a i t s ..............
M e t h o d o l o g y ................................
Procedures I:
Clustering the Traits . . .
Procedures II:
Grouping the
D e p a r t m e n t s ..............................
S u m m a r y .....................................

32
33
38
45

THE R E S U L T S ............

49
52
53

The Clusters of T r a i t s .....................
The Groups of Departments . . . . . . . .
Descriptions of Departmental Groups
. . .
S u m m a r y .....................................

53
67
73
80

D I S C U S S I O N ........................................

82

Problems from the M e t h o d o l o g y ............
Problems from the D a t a .....................
Trait-Clusters ..............................
Department Groups
.........................
Relation of Department Groups and
C o l l e g e s ...................................

82
87
88
91

iv

92

Chapter

Page
Department Typology
....................... .
96
The Two General Classes of Department
T y p e s .......................................
99
The Six Basic Types of D e p a r t m e n t s ............ 101
Subject Matters of the Departmental
T y p e s .......................................... 104
S u m m a r y .......................................... 107

VI.

C O N C L U S I O N S ......................................... 109
Evaluation of the Procedures and Data
. . .
Attainment of Objectives .....................
Significance ..................................

BIBLIOGRAPHY

..............................................

109
110
113
119

APPENDICES
A.

DEPARTMENTS THAT MADE UP THE POPULATION
OF I N T E R E S T ........................................ 123

B.

DEPARTMENTAL TRAITS THAT MADE UP THE DATA . . .

C.

PROGRAMS AND PROCEDURES THAT MADE UP
THE M E T H O D O L O G Y ................................... 139

D.

CLUSTERS OF DEPARTMENTAL TRAITS

V

L

................

127

141

LIST OF TABLES
Table

Page

1.

Alphas for T r a i t - C l u s t e r s .....................

54

2.

Department-Group Alphas

........................

67

3.

Department Groups:
Scores on TraitClusters .
.....................................

74

4.

Relation of Trait-Clusters and Departments

90

5.

Departments in Each Group that Belonged
to the Same C o l l e g e ....................

6.

7.

. .

. . .

Relation of Department Groups and Colleges,
From College Perspective .......................
Department Typology

............................

93

94
100

LIST OF FIGURES

Figure

;•

L

Page

1.

Correlation of Trait-Clusters

...............

55

2.

Department-Groups Correlations

...............

68

vii

CHAPTER I
BACKGROUND

In most universities,
structural unit.

the department is the basic

Most faculty members belong to a depart­

ment and depend upon it for their initial hiring, promotion,
salary raises, tenure award, research opportunities,
various perquisites (travel allowances,

and

clerical aides, etc.).

Through its m e m b e r s , the department manages the preparation
and execution of instructional courses and degree programs
(Joughin,

1971; Clark,

1971).

It follows that to understand

the university, one must also understand the department,
both as a unit of the structure of American higher education
generally and as a unit of specific university structures.
However,

unfortunately little research or discussion was

focused on the department before the 1960s.

This study has

been undertaken in light of these considerations and with
the intention of contributing to the fund of information in
this important area of study.
A university department can be defined as an orga­
nizational unit whose members are practitioners of a single
academic discipline, technology, or area of study; whose
purpose is to represent its members'

interests concerning

2

(1) planning and implementing academic courses and degree
programs related to that discipline, and
members'

(2) managing the

professional careers, professional certification

and socialization,

and relevant personnel matters.

The academic department appeared in the American
colleges as early as the late eighteenth century.
department developed during the nineteenth century,

As the
it

usually came to reflect a single academic discipline,
ject matter or area of inquiry.

sub­

During the past one hundred

years, both the number and variety of departments have
greatly increased as university founders and administrators
responded to pressures generated by changes in the intellec­
tual, technological,
American university
1968; Rudolph,

social, and political contexts of the
(Anderson,

1962; Vesey,

1968; Brubacher and Rudy,

1965).

The development of the department should be analyzed
in relation to other departments in American culture.
However,

the reasons advanced by university founders and

reformers for their adoption of such and such a specific
departmental structure have rarely been recorded or commu­
nicated to later investigators.

Historians can only muster

ex p os t f a o t o speculations for the causes of departments'

development, even though the actual events— the dates of
establishing certain departments— are more or less factually
known.

Vesey

(1965, pp.

267-8)

states the problem thus:

3

The internal structure.of the American
university rapidly acquired the shape which
in most respects it would maintain from that
time forward [ 1 8 9 0 - 1 9 1 0 ] . . . . But exceed­
ingly little direct evidence may be found on
decisions involving the ba sia shape of the
rapidly emerging academic structure.
The
most fundamental assumptions were not being
articulated by those w h o were acting upon
them.
Whatever the direct evidence,

it is clear that the

modern university has continuously changed its departmental
structure by institutionalizing many of the intellectual and
technological innovations
seventeenth century.

that have occurred since the

T. N. Clark

(1968)

suggests four

processes by w hich this institutionalizing has occurred:
organic growth of new professions which led to new d i s c i ­
plines being brought into the university;
(division of labor)

differentiation

among existing disciplines;

diffusion

of new knowledge from one institution to another;
process combining the other three.

and a

This process of insti­

tutionalization profoundly altered universities'

structures

both in America and Europe during the nineteenth century and
continues in the present day.

Differentiation and synthesis

of subspecialties has continued, producing departments of
bioengineering, biochemistry, biophysics, plant pathology,
parasitology,

and so on; and the development of new tech­

nologies and professions has also continued, producing
departments of social work, police administration,
science, computer science,

and so on.

food

4

The differentiation,

diffusion, and accretion that

have led to the creation of new departments have operated
in areas besides the traditional disciplines.

In one area,

several long-standing occupations— lawyers, doctors, m i n ­
isters,

school teachers— "professionalized" themselves

before 1900 and encouraged universities to establish schools
for selecting and training new members of the profession
(Jencks and Riesman,

1969, Chapter V).

In another area,

many universities responded to the needs of the rapidly
developing and expanding industrial economy after 1865 by
creating new schools or departments for technological
studies.

Departments became sources of new knowledge and

professional personnel not only in relatively traditional
disciplines such as chemistry and physics, but also in new
areas of applied knowledge, such as agriculture and engi­
neering

(Halsey,

1960).

As the industrial revolution grew

into the American "organizational revolution" universities
added new departments,

schools, and colleges to represent

the new technologies of business,
Halsey

(1960)

finance,

and management.

saw the department filling a new role in the

modern era:
In this new technological society educational
institutions are expanded not only to exercise
research functions but also to play a central
role in the economy and the system of stratifi­
cation as agencies for selection., training, and
occupational placement of individuals (p. 463).

5

The American tendency to incorporate technological
studies into the structure of the universities and colleges
differed from the European response to similar pressures.
Throughout western Europe, the French Ecole Polytechnique
became the "prototype of all colleges of higher technology"
and "the production of manager-technologists was entrusted
to polytechnics which in the course of the

[nineteenth]

century acquired the rank and prestige of universities"
(Ashby,

1958, p.

472).

This European solution to the p rob­

lem of technical education resulted in a division of labor
in the academic world:
sciences and arts;

the universities handled the "pure"

the technical institutes the new applied

sciences.
In England,

the universities incorporated technology

into their structures,
the Americans,

as in America.

But in contrast to

the English did not, until recently,

give

adequate support to technological studies either with money
or by preferring university-trained personnel for managerial
positions in industry

(Ashby,

1958).

American universities adopted technology w h o l e ­
heartedly,

sometimes creating new schools and colleges

to separate the so-called professions from the so-called
dis c i p l i n e s , 1 but almost always ultimately giving the

P e n c k s and Riesman (1969, pp. 236-250) consider
graduate training in the disciplines to be as much a pro­
fessional operation as the training in engineering, law,
and so on.

6

technologically-oriented units full status as departments
with the right to offer courses and degrees and to develop
their own f aculties.1
The general pattern of departments in the American
universities was pretty well set b y 1900, but this pattern
has undergone continual change and innovation and continues
to change in the present day.

Universities have continued

to establish new departments representing new knowledge and
new relationships among the established disciplines,

as well

as meeting the demands for new training and research arising
from social or economic changes.

As their contexts change,

departments also change their characters and their relation­
ships with other departments,
sity administration.

other units,

and the univer­

Some of these changes are exemplified

by the shift in the relative status of classics departments
vis-a-vis chemistry departments;

during the past fifty years

the former lost its prestige as the epitome of humanistic
scholarship while the latter derived new prestige following
World War II and Sputnik.
In the present day a trend toward creating another
kind of innovative department has arisen from the desires
of faculty and students to focus upon a problem

(or group

of p r o b l e m s ) , an area of study, or an area of application;

xThe few technical institutes set up as separate
institutions— Massachusetts Institute of Technology,
Rennselaer Polytechnic, and others— were exceptions to the
general pattern.

7

rather than upon a single discipline or technology.

Several

disciplines and technologies may be combined in this kind of
interdisciplinary department in order to seek a joint solu­
tion to a broad problem or to illuminate a broad issue.
Departments of ethnic, racial, urban,

regional or ecological

studies reflect these current interdisciplinary interests.
This historical development has produced a contem­
porary American university that is extremely heterogeneous,
incorporating several fields of study that are treated by
separate institutions in most other countries.

The depart­

ments included within this peculiar type of university— or
within its component colleges— represent single disciplines,
combinations of disciplines,
technologies,

technologies,

combinations of

and areas of application for both technologies

and d i s c i p l i n e s . 1

'From the departments listed in the catalogs of two
major state land-grant institutions, the University of Min­
nesota and Michigan State University, one can form at least
ten broad categories of subject matters that are treated by
separate departments.
These categories and some represen­
tative departments are indicated here.
(These examples are
for illustration only, and various departments could be
argued to fit into other categories.)
1.

Single disciplines:
Humanities:
Social Sciences:
Biological Sciences:
Physical Sciences:
Single disciplines:
Humanities:
.
Social Sciences:
Biological Sciences:
Physical Sciences:

Types of Subject Matter
Tools_____ Abstract
Concrete
Languages Philosophy Literature
Statistics
History
Statistics
Statistics Math
Experimental

Botany
Chemistry

Quasi-Experimental
Linguistics
Sociology
Psychiatry
Astronomy

Besides this great diversity in their subject
matters, departments have developed wide variations in their
other essential aspects— governance, membership, operational
functions, resources,

and objectives.

With respect to

governance, departments have different decision-making
structures

(formal or informal, perceived or not perceived)

as well as different styles of governance and administration
(Dressel et a l . , 1970; Corson,

1960, pp.

85-94).

With respect to departments' membership,

several

studies have suggested some of the ways in which departments
may differ.
Freedman

Selvin and Hagstrom

(1960)

and Bereiter and

(1967) studied the relation of students' liberalism

to their areas of study.

Vreeland and Bidwell

(1966) used

student-faculty interaction as a dimension for classifying
departments.

Currie, Finney, Hirschi, and Selvin

found that the numbers of students

(1966)

"realistically consid­

ering" becoming a college professor varied from one area

2.
3.
4.
5.
6.
7.
8.
9.
10.

Mixed disciplines:
Art History, Music (theory) Religion
Sub-disciplines:
Child Psychology, Microbiology
Combined sub-disciplines:
Biochemistry, Geophysics
General studies:
Humanities, Natural Sciences,
Social Sciences
Quasi-disciplines:
Management, Accounting,
Administration
Technology:
Architecture, Library Science, Civil
(etc.) Engineering
Combined disciplines and technologies:
Biometry,
Ecology, Agricultural Economics
Practical arts:
Studio Arts, Journalism, Music
Areas of application:
Afro-American Studies, Home
Economics, Criminal Justice, International Relations,
Urban Planning.

9

of study to another.

Goff and Wilson

the areas of humanities,

(1971)

suggested that

social studies, natural science,

and professional studies were four different "faculty cul­
tures" with different attitudes toward certain relevant
topics *

By implication, these studies— even though often

focussed on areas of study rather than departments— lead
toward the theory that different departments attract d i f ­
ferent kinds of members, both as students and as faculty;
that departments have different value "climates" or
"personalities";

and that departments have different kinds

of effects upon their members.
Many other characteristics of department m e m b e r s —
both as individuals and as groups— also deserve to be
studied for the light they might shed on the essential
aspects of departments.

Faculty size, for example, has

been cited as significant for explaining differences between
departments both within and between universities
1964; Haas and Collen,

1963).

(Murray,

The size and characteristics

of the student members and the non-academic staff have not
been analyzed in relation to other departmental aspects,
even though they clearly have profound effects on how
departments actually function.
Departments'
facets:

basic operational functions have two

the kind of activities undertaken, and the manner

10

of performing those a c t i v i t i e s . 1

Before the academic

department can be properly studied,

its functions must be

adequately defined in their true extent.

Ordinarily,

only

one or a few departmental functions have been the topic of
research or discussion.

For example,

the department's role

in recruiting and socializing new members for the profes­
sions

(academic as well as "applied") has not generally

received as much attention as its role in research and
teaching

(Gottlieb,

1961; Rosen and Bates,

1967).

Even without a discussion of the variation in
departments'

o b j e c t i v e s 2 or resources,

the point has been

amply illustrated in this discussion that departments in
American universities are not only extremely heterogeneous
as a group, but that their significant differences occur in
several aspects besides their subject matters.
matter in another way:

departments have several essential

aspects of which subject matter is only one.
functions, members,

resources,

must also be studied.

To put the

governance,

Departmental

and objectives

All these aspects are complex topics

^ a s i c departmental functions may include all the
operations involved in admitting and certifying students,
teaching, preparing courses and programs, researching,
hiring and rewarding teachers, serving off-campus clients,
and so on.
The manner of functioning includes such var i ­
ables as class size, teaching techniques, admission
requirements, tenure requirements, and so on.
2See Vreeland and Bidwell,

1966.

11

that must be analyzed into their components, but must also
be related both to one another and to the university at
large.

Only when many studies of these topics have been

achieved can we begin to understand the department and its
parent institutions.
In spite of the heightened interest in institutions
of higher education that followed the student demonstrations
in the late 1960s and the scarcity of money in the early
1970s, analysis of the department as a structure or as a
kind of formal organization is a task that has not attracted
many researchers.

But students,

faculty, and funds cannot

be adequately understood as resources for the university
until they can be clearly related to the basic operating
units of an institution.
analysis,

Program budgeting, cost-benefit

and any serious evaluation of institutional e f f i ­

ciency will continue to founder unless the actual workings
of the university departments are fully understood.
Several recent studies of contemporary y outh—
Kenniston's,

for example— have illuminated some of the

social forces that seemed to have created an on-going crisis
on many campuses in the recent past.

However,

activist st u ­

dents were not only the products of certain family environ­
ments and peer r e l a t i o n s h i p s , but also in many cases were
members of university departments.

S tu d en t protest,

opposed to protest by youth generally,

as

cannot be understood

completely unless the analysis includes the academic context

the structures and functions of the university and its units.
Possibly,

the dysfunctions and counter-productiveness so

often ascribed to departments have also been a cause i>f
student alienation.
Clark

(1971)

suggests that academic society be

understood in terms of the interplay of four concepts:
community, bureaucracy,

interest group, and profession.

When he considers the department in relation to professions
and professionalization, Clark finds a problem peculiar to
the academic profession in the fact that its professionalism
operates in terms of the separate disciplines and is orga­
nized into separate departments.

He states

(p. 38),

"Disciplines and departments tend to have a centrifugal
force.

. . . The effect is a fracturing of the university

and college."
An opposite centripetal force has also affected the
university's structure.

Vesey

(1965, p.

311) cites "two

countertendencies at work in American higher education:
fragmentation and centralization."

He considers the devel­

opment of a central administrative structure to be both a
result of the centralizing tendency,
to the fragmenting tendency.

as well as a response

He concludes that:

Bureaucratic administration was the structural
device which made possible the new epoch of
institutional empire-building without recourse
to specific shared values [or, among the disci­
plines, shared methods and principles].
Thus,
while unity of purpose disintegrated, a u n i ­
formity of standardized practices was coming
into being (p. 311).

The fundamental organizational problem facing the
emerging American university was

(and still is)

to create

new types of academic units that could adequately represent
the needs of an academic discipline, profession, or tech­
nology;

to create enough of such units to cover the full

spectrum of human knowledge

(or as much of it as an insti­

tution's resources could afford);

to grant them enough

autonomy to perform their functions well; and then to bring
them under enough control by a certain administrative
bureaucracy for the maintenance and advancement of the
institution

(Corson,

1960, pp.

33-42).

One of the organizational patterns developed to
provide central control over departments is a college that
includes all or most of the academic disciplines

(called

the college of arts and sciences, or something s i m i l a r ) .
Another common pattern organizes the departments into
several colleges or divisions on the basis of common subject
matters,

such as four divisions that represent the humani­

ties, social sciences, physical sciences,

and biological

sciences.
Whether the organization that subsumes the depart­
ment involves one college or several, there are problems.
Spurr

(196 8, p. 11)

found the single college of science and

arts to be "too large, too complex, and archaic in its orga­
nization."

As for the reorganizations mos t frequently ’

encountered, he found that the division has had widespread

14

acceptance at graduate levels but less at undergraduate
levels, and that the sub-college

(residential or theme

college) may be relevant or appropriate only under limited
circumstances.
In a divisional organization, departments are
grouped according to assumed similarities in their subject,
matter, but little if any evidence supports such assumptions.
As has been pointed out above, departmental subject matters
are of more than one kind;

they may be disciplines,

ogies, areas of study or application.
categories

(such as the disciplines)

technol­

The subject matter
are also heterogeneous,

representing different segments of several continuums:
example,

for

in methodology, a continuum from the "purely"

experimental studies to the quasi-exp e r i m e n t a l ; in concepts,
from the abstract to the concrete;
ical to the practical.

in form,

In addition, even departments that

have subject matters obviously similar
of French and Spanish)

from the theoret­

(such as departments

may have different governance struc­

tures, administrative styles,

and functions.

There is

little real similarity among departments, no matter what
the appearances.
The question should be posed, whether another system
of grouping departments might not be more efficient than
either the single-college, divisional,

or multi-college

systems almost universally employed by contemporary American
universities,

"efficient"

in the sense of better utilization

15

of r e s o u r c e s .

But the concept should not be limited to

those aspects of departmental operations which can be
precisely defined and measured.

Nevertheless,

the un i ­

versity's efficiency can be better assessed and thereby
increased if a department can be compared with other
departments that it most closely resembles.

Obviously,

comparison of unlike entities is invalid, but the de p a r t ­
ments usually grouped together are not necessarily similar
in their essential aspects.
Some objections to divisional organization have
been expressed by Kim

(1966).

"Divisions" are necessary for the sake of
administration.
The main issue— assuming that
there is cause to raise an issue— is the ten­
dency of administrative compartmentalization
to imbue its participants with a psychology
of separation. . . .
Obviously, the common identity factor, which
is the basis for the arrangement of particular
departments into a division, is as sensible as
calling politics a science.
For example, some
of the academic disciplines which make up the
division of humanities, under certain circum­
stances, can be more effective in the division
of social sciences, perhaps at times in the
division of science.
The effectiveness is not
dependent on the definition of a discipline as
much as it is contingent upon the purpose of
the endeavor. . . .
There must be a persistent effort to appraise
the value and function of the division.
And
the appraisal should produce one of the follow­
ing recommendations:
(1) the exclusion of some
departments; (2) the inclusion of other depart­
ments; or (3) determination of common areas of
pursuit designed to maximize the realization of
purposes (pp. 428, 430, 431).

16

Jencks and Riesman

(1969, p. 523) cite a "need for

much greater flexibility in the grouping of strictly aca­
demic skills and expertise."

Their analysis is so relevant

to the present discussion that their statement deserves to
be quoted at some length.
A discipline is at bottom nothing more than an
administrative category.
The various subdisci­
plines within biology or history or psychology,
for example, have only the most limited intellec­
tual relationship to one another, and the same
is true in every other field.
They are grouped
together mainly because the men working in them
went through the same sort of graduate program
and have some residual feeling of common identity.
A good deal of ingenuity has, it is true, been
devoted to the rationalization of these tradi­
tional ad hoc arrangements. . . .
The issue here is not specialization versus
generalization.
The issue is whether one way
of aggregating specialized skills is better
than another. . . .
A t any given moment there would be some advan­
tage to regrouping the various subdisciplines
into new combinations, simply because the new
units would be less hallowed by tradition and
more subject to criticism and ad hoc m o d ifica­
tions.
But over the long haul there is no
reason to think new combinations w ould have
any significant advantage over the old ones.

Abolishing old departments and establishing
new ones must, therefore, be a continuing
process, like Jefferson's revolution every
twenty y e a r s . . . . We are not urging the
superiority of one combination over another,
only the advantages of different combina­
tions . . . .
There is a good
the departments'
them to smaller
lectual work is

deal to be said for curtailing
present powers and distributing
groups.
The real unit of intel­
the subdiscipline, which usually

17

has only one or two representatives on a small
campus and seldom more than half a dozen even
on a big one.
These subdisciplines are eva­
nescent, arising and disappearing over the
lifetime of a single faculty generation.
They
fertilize one another in unpredictable ways,
and the curriculum should be sufficiently
flexible to accommodate this tendency (Jencks
and Riesman, 1969, pp. 523-527).
MacMahon has stated a similar viewpoint
1960, p.

(in Corson,

34):

The inherited departmental structure must not
be allowed to interpose an element of rigidity
that works against the advancement of knowledge.
A decisive test of the university's organization
is the degree to which it insures a continuing
re-examination of the specific departmental
structure that it has inherited.
Spurr

(1968)

for universities:

suggests a two-dimensional organization

in one dimension, departments would

handle courses and programs for "concentration" requirements,
while in the other dimension,

inter-disciplinary faculty

groups would handle courses and programs for "distribution"
requirements

(a scheme somewhat similar to the current

systems at the University of California at San Diego and
at Santa Cruz and the University of Wisconsin at Green B a y ) .
However,

S p u r r 's proposals would lead to problems because,

first, he seems to consider chiefly the department's under­
graduate functions;

and second, he considers only one

departmental function— the courses and programs offered—
while ignoring other crucial functions such as student and
faculty recruitment,

research,

and faculty rewards.

The

reward function— promotions, raises, tenure— has in fact

18

produced severe tensions between the two kinds of faculty
groups

(disciplinary/departmental vs. interdisciplinary/

collegiate) where such a system has been adopted.
Another problem in a two-dimensional system is that
departments are interrelated only by means of their curriculums, while other aspects are ignored.
Clark Kerr

(1970, pp.

119-121), discussing univer­

sity governance in a broad sense, suggested a "pragmatic,
pluralistic strategy"

that can be applied equally well to

departmental organization:
No single solution is possible--situations vary
from campus to campus and within a campus from
one major function to another.
Nor are perfect
solutions likely. . . . Nor are permanent solu­
tions likely. . . . There can be no clear pre f ­
erence for one solution versus another solution
on principle, given the nature of the academic
institution. . . . A practical, pragmatic
approach seems more sensible. . . . A new series
of consensuses is needed— not one overall "best"
solution, not one single preferred form.
Variety
in solutions should match the variety of situa­
tions and this is almost infinite.
Ikenberry

(1972) proposes:

task oriented units as a supplement or as an
alternative to the academic department [with]
a more specific and restricted mission definition
[relative to the departments] and a greater use
of part-time, jointly appointed staff members
from several disciplines (p. 31).
Perhaps the university needs a pluralistic, m ulti­
farious system of departmental organization, in which
departments could be grouped in various w a y s , according to
the needs within specific institutional contexts.

Groups

of departments could be formed in several ways according to

19

various aspects of subject matter— organized into divisions
or dichotomized into theoretical-practical, experimentalnonexperimental,

analytical-synthetical,

and so on.

One

could also form several kinds of groups according to various
departmental functions— all, several, or one
fessional socialization,

curriculum,

(such as pr o ­

faculty career ma n a g e ­

ment, productivity in degree and credit hours, and so on).
In addition,

the structures and styles of governance could

be used to form groups of departments according to such
traits as authoritarian-democratic,

formal-informal,

and

so on.
Departments could be grouped in one w a y for curric­
ulum matters,

in another way for budgeting,

number of ways for other purposes.
ture could be pragmatic, ad hoc,

and in any

The organization struc­

flexible, protean.

It

could adapt readily to disciplines and subdisciplines that
are constantly changing,
put it.

But,

"evanescent," as Jencks and Riesman

the system would still be based on concrete

needs and empirical evidence.
Such a system might be able to replace the current
chaos of university structure— the maze of centers,
tutes,

colleges,

schools,

insti­

senates, unions, offices, bureaus,

and casual interest groups— with a unified structure that
can accommodate complexity and provide for orderly growth
and change.

The university needs a unified theory of

organization that can turn it around like the Copernican

20

revolution and can free it from the present absurd system
in which units proliferate like Ptolemaic epicycles.

Summary
1.

The department is the university's basic

structural unit, and will remain so in mos t institutions
for the foreseeable future.
2.

The department has developed during the past

100 years into a type of organization that is highly
heterogeneous, exhibiting great diversity in every aspect.
Departments have continued to change in many ways since, the
general pattern of American universities was established,
roughly around 1900.
3.

Departments have at least six essential aspects:

subject matter,
and objectives.

resources, governance, members,

functions,

All of these aspects m ust be studied in

order to achieve understanding of the department and the
institutions of which it is a part.
4.

The organizational structures— colleges, schools,

divisions— that have been established to interrelate and
manage the departments within a university have usually been
based on assumed similarities and dissimilarities in the
departments'

subject matters.

These traditional structures

are usually inadequate to handle the reality of departments'
operations because the structures have not been based on
consideration of all the essential aspects of departments.

21

These structures may be at times dysfunctional because they
group together departments that are in fact dissimilar in
ways that are crucial in a given situation.
5.

Universities should consider new organizational

structures in which all departmental aspects can be ade­
quately related.

New structures could take several hitherto

untried forms such as temporary,

ad hoc groupings, or over­

lapping, multi-dimensional systems.

CHAPTER II
THE PROBLEM

In Chapter I, the discussion was focussed on a
general problem in higher education,

that the structures for

organizing departments within a university are inadequate.
This chapter will focus upon some research studies relevant
to the classification of departments, some deficiencies in
this research,

and some possible means to overcome such

deficiencies.
At a recent seminar on university organization and
governance,1 among the twelve well-known researchers in
higher education who were contributors or c o m m e ntators, only
two made statements directly relevant to the department,

a

strange situation considering the importance of departments
in university operations.

One of the participants referred

to "subunits" in his highly abstract discussion of univer­
sity decentralization; but even he was not really interested

xThe Invitational Seminar on Restructuring College
and University Organization and Governance, sponsored by the
American Association for Higher Education with the support
of the Kellogg Foundation, February 28-March 2, 1971.
Con­
tributed papers and commentaries reprinted in the J ou r n a l of
H ig h e r Ed u ca ti o n, 42, 6 (June, 1971).

22

23

in the functions of departments.

Omitting explicit

references to the department from a discussion of re­
structuring the university is almost as incomprehensible
as overlooking the bricks in restructuring a chimney.
The problem of research on university departments
is not only that so little of it is actually being done, but
also that so much of what is done is not really adequate.
Most research on departments involves a conception of a
department that is too limited, a research design and method
that is too simple, and data that are too imprecise.
Researchers in higher education have suggested
various bases for classifying departments besides the
traditional separations based on subject matter or on a
distinction between theoretical and practical purposes.
Some of these bases include forms of governance,

formaliza­

tion and decentralization of governance, educational goals,
and intra-departmental social interaction.

Some of these

studies will be discussed in this section.

(Other studies—

Bereiter and Freedman,

1967;

Selvin and Hagstrom,

1960— have

categorized academic fields of study according to attitudes
of students in the various fields, but extrapolation from
fields to departments is risky.)
Haas and Collen

(1963)

categorized departments

according to high, medium, or low levels of formalization
of procedures concerning faculty hiring, evaluating, and
dismissal, basing their analysis on interviews of eighty

24

department chairmen in one university.

Their study is of

limited value because they did not consider certain relevant
variables that could confound the findings:
of department chairman,

e.g.,

"style"

duration of traditional practices

in older d e p a r t m e n t s , numbers of departmental faculty and
students,

and so on.

Also,

their findings could be gener­

alized to other universities only with many qualifications,
if at all;

the same comment applies to almost all studies of

universities,
Moran

including the present one.
(1971) measured the decentralization of dec i ­

sion making from a central administration to a certain sub­
unit,

in eleven universities, using a structured interview.

He developed an index of decentralization for ranking or
grouping university subunits.

Although

the type of subunit he studied,

one can

he did not specify
assume from his

interview devices that it was a college or professional
school rather than a department.

As in the previously cited

studies, the confounding variables left uncontrolled or
unexplained tend to vitiate his findings.
amount of outside funding,

the prestige

For example,

the

of the subunit,

the

character and administrative style of the subunit's chief
administrator could all affect the decentralization of deci­
sion making.

Moran correctly assumed that the subunits

chosen for study should all be of the same type, but he
was not correct in assuming that the same kind of subunit
at eleven different universities w ould be comparable.

25

Murray

(1964) defined five categories of departments

that he hypothesized were also developmental stages in the
history of departments.

He differentiated the five types

of department by their governance structures and styles.
His informal analysis of 22 diverse universities suggested
a "discernible pattern of departmental development w hich was
intimately connected with university size, general campus
administrative complexity,
(p. 228).

and institutional prestige"

He did not describe his methods of interpretation

and analysis,
his findings

and therefore one cannot accurately evaluate
(he tells nothing of his selection p r o c e d u r e s ) ,

but one could easily doubt that the 22 institutions were
actually similar enough to permit comparison of their
departments,

to say nothing of the great diversity of

departments that he collapsed into five types.
Vreeland and Bidwell

(1966)

interviewed one hundred-

twenty-seven faculty members in 22 departments at one univer­
sity, and then categorized departments according to faculties'
goals for undergraduate education and the social interactions
within departments.
technical, moral,
interaction:

They defined three kinds of goals:

or mixed;

low, medium,

and three levels of departmental
or high

(in terms of faculty

interest in students and student-facuity int e r a c t i o n ) .
However, their study also fails to control for certain
confounding variables;
tories,

e.g.,

in departments that use labora­

the form and meaning of student-facuity interaction

26

is different from those that use chiefly discussion sections.
Also, the study depends on the faculty's perceptions of the
departmental goals and the s t u d e n t s ' perceptions of personal
relationships, but— since evidence suggests that certain
fields attract certain personality types— one could question
whether student-facuity interactions are comparable across a
diverse group of departments.
These studies illustrate that departments can be
classified according to other departmental aspects besides
subject matter.

They focus on relationships that indicate

possibilities for further research, but they are of limited
use because they focus only on a limited conception of
departmental aspects.
be a wrong approach,

A limited focus may not in itself
since university departments are so

complex and diverse and our present state of knowledge about
them is so inadequate that no single system of categories or
single organizing structure can handle the actual concrete
situation.

Nevertheless,

adequate studies of departments

should consider as many of the departmental aspects as
possible.

This is not an easy task,

since no theoretical

constructs are available to consolidate the data describing
departmental functions.

Researchers face the problem of

coping w i t h large amounts of diverse data.

Multivariate

methods have not been usually applied to studies of depart­
mental functions and c h a r a c t eristics, although they would
seem to be highly suitable.

27

Another problem apparent in studies of universities
and their departments arises from the nature of the data,
which is usually

"soft"— opinions, self-reported perceptions,

attitudes, and so on.

This situation is largely unavoidable,

as in mos t social sciences.

However,

the data in studies of

departments have been almost entirely on the level of
opinions or perceptions.

This situation seems unnecessary,

considering the large amounts of data available from various
university offices

(registrar,

institutional research).

In studying universities, generalization is also
a problem since each university— at the present state of
theory--appears to be an entity separate in itself.

If one

were to consider most of the factors that apply to univer­
sities

(size, type of control, prestige, financial resources,

and so on) one could construct a matrix with many empty
cells.

One of the causes of this situation is that most

universities have developed not according to theories of
education or social needs, but in response to a specific
historical situation.

The concrete historical setting

affects a university's essential aspects so that they work
together differently from the way the same aspects work
together in other universities.

Each university is a unique

combination of traditions, structures,
community pressures,
setting.

legal frameworks,

and clienteles, in a specific physical

As a result, valid generalizations are probably

28

more difficult to make about universities than about other
types of institutions.
The present study has attempted to overcome some of
these research inadequacies,

or at least to point the d i r e c ­

tion which should be taken in such an attempt.

This attempt

has been embodied in the use of a certain kind of data and a
certain kind of method, both of which are appropriate in
terms of a certain conception of a department.
In the present study,

a university department has

been conceived as a complex organizational unit with six
essential aspects:
governance,

subject matter, members, resources,

functions,

and objectives.

All of these aspects

must be accounted for in a truly adequate theory of the
department.
Therefore, in this study an attempt has been made
to include as wide a selection as possible of departmental
descriptor variables,
aspects.

to reflect several of the departmental

However, the traits chosen have for the most part

been related to departmental functions, such as course
offerings, student credit hours produced,
and so on.

teaching modes,

There are two reasons for this selection.

First,

a department's functions express the basic character of a
department— both as a general type of organization and as a
specific example.

Members, resources,

and governance are

also all equally as necessary to a department's existence
as its functions, but knowing about those three aspects of

29

a department does not tell one muc h about its functions,
while knowing about its functions does give one some basis
for making inferences about the other aspects.

The relation

of functions and objectives is somewhat different.
in a sense two conditions of the same entity,

They are

since func­

tions are the embodiment or actualization of the objectives;
the one involves physical actions,

the other involves ideas.

A problem arises because the relation of functions and
objectives may be perceived differently;

different persons

may perceive the same action as an embodiment of different
obje c t i v e s .
The intention of avoiding this problem of perceiving
objectives is a second reason for selecting chiefly data
describing departmental functions,

for the present study.

The data in much of the previous research have been drawn
from students'

attitudes and faculty opinions.

Such data

are appropriate for studies of such topics as the style of
departmental governance,
tions on its members,

the effect of departmental func­

or the prestige of departments.

But

such data necessarily relate to someone's opinion about a
department,

and not to the department's functions per se.

The present study seeks to avoid some of the inade­
quacies of research based on opinions and perceptions by
selecting data chiefly from the functions of a department,
and objective in form.

30

The present study has also sought to use an
appropriate method for this kind of data.

Such a method

would have to be capable of handling large amounts of data
for w hich no objectively valid criteria for categorization
are available.

The method must facilitate grouping

the data

in terms of relationships that are inherent in the data but
are not easily revealed by ordinary processes of observation
An appropriate method was found in a form of factor
analysis which employs the techniques of correlation and
multiple-groups analysis to achieve clusters or groups of
subjects, as will be described at greater length in the
succeeding chapter.

This method has the additional merit

of handling the data in terms of a specific situation, which
is appropriate for the conception of the university that has
been adopted in the present study, namely,

that until a

general theory of the university has been developed, each
university should be treated as a unique organization, a
specific historical event in a specific physical setting.

Summary
Most studies of university departments have been
inadequate.

They have been designed in terms of a limited

conception of the department's characteristics and functions
and as a result the designs have not had proper controls for
possibly confounding variables, and have not employed a
sufficient number of variables.

The designs have not been

31

appropriately complex for the subject.

The studies have

employed data that are chiefly opinions and perceptions,
ignoring available objective data.

The studies are not

generalizable beyond their population because in the present
state of our knowledge about universities, we cannot assume
that universities or various types of universities are
actually comparable.

Each university must be treated as

ultimately a unique institution until sufficient evidence
has been accumulated to permit a valid definition of un i ­
versity types.
The present study proposes to seek a more adequate
form of research by

(1) selecting chiefly data that

describes the functions of a department and is expressed
in an objective form; and

(2) employing a method of analysis

that can handle large amounts of disparate data and reveal
the relationships that are inherent in the data.

These data

and methods are appropriate for a conception of the depart­
ment as an organization that has several essential aspects
to be accounted for, and a conception of the university as
an organization that is essentially unique, a specific
entity in a specific context.

CHAPTER III
OBJECTIVES,

DATA, PROCEDURES

Objectives
The objectives of this study were to analyze
objective traits describing departments'

functions in order

to determine dimensions in the traits upon which departments
differ, and then to group the departments on the basis of
those dimensions, by means of a process of clustering and
multiple groups a n a l y s i s . 1
A basic question of this study is whether,

if

departments are grouped according to similarities in their
functions,

the resulting set of groups will differ from a

set of groups based on similarities in the subject matters
of the departments.
The success of the study can be evaluated according
to whether the groups established by this process are rea­
sonably precise, meaningful,

and useful.

xThe term "trait" has been used in this study to
indicate the descriptive variables or characteristics of
departments.
It was used, first, because it avoids the
possible confusion arising from use of "variable."
When
the departments are grouped according to their traits, the
departments function as the "variables" and the traits as
"subjects" in a multiple-groups analysis.
Second, "trait"
is shorter and therefore easier to use than "characteristic."

32

33

In this context, precision means that the groups
exhibit internal consistency and external discreteness.
M e a n i n g f u l n e s 8 means

that the groups can be described in

understandable terms, and that the groups reflect the real
situation in the university as interpreted by an experienced
observer.

Utility means that the groups could provide the

rationale for an organizational structure of the departments
In addition to evaluation of the study's results,
the efficiency of the procedures should also be evaluated.
E f f i c i en c y

in this context means that the time, effort, and

money expended on the study can be justified by the value of
the results.

Also, the study can be judged to be more or

less h e u r i s t i c to the extent that it implies further studies
or provides concepts that can be used as unifying agents in
understanding the university generally.

Data:

Population and Traits
The population of interest for this study consisted

of 79 of the academic departments of Michigan State Univer­
sity in 1969 to 1970.

A department was defined as

administrative unit of the university,

(2) to which person­

nel are assigned for salary and career management
t e n u r e ) , and

(1) an

(promotion

(3) which possesses and exercises authority to

offer courses of instruction.
Five departments at this university were excluded
from the study:

the four departments of the College of

34

Education, because their figures for enrollments, budgets,
e t c . , are reported in an amalgamated form as college totals
rather than separately for each department;

and the Depart­

ment of Family Ecology, because it offered no courses in
1970-71.
A list containing the departments studied, grouped
by the colleges to which they belong, is shown in Appendix A.
The numerical order of departments in that list was used to
identify departments throughout this study.

For the most

part, the clustering process was accomplished by referring
to departments'

numbers, without further identifying them.

This "blind" approach was used to overcome the investiga­
tor's preconceptions of which departments belong together
and to avoid operating in terms of subject matter designa­
tions .
The departmental traits used in this study were
obtained from reports or files prepared routinely at
Michigan State University,

and the reports of the graduate

education prestige published by the American Council on
Education.

The traits are listed in Appendix B.

The main source of traits were the MSU Office of
Institutional Research,

the MSU annual budget report,

financial reports prepared by other MSU offices,

and a

file of current faculty data maintained by the MSU Provost's
office.

The traits totaled 167.

35

The general purpose of this study implies that the
departments should be grouped on the basis of data that is
objective and empirical rather than subjective and a priori,
and grouped on the basis of the departments'
rather than their subject matters.

functions

The traits in this study

were appropriate data in terms of this purpose,

since they

were all at least relatively objective or as objective as
possible in the actual present situation.

The department

traits were objective and verifiable to the extent that they
expressed observable information.

Some of the figures were

verifiable by reference to published documents such as the
MSU catalogs,

schedules,

and budget.

Several traits

appeared in more than one report, e.g., student credit
hours;

thus, the consistency of the reports further verified

the figures reported for such traits.

Many of the traits

were recorded directly from registration figures and thus
were obtained independently of the departments to which they
refer.
It might appear that the total of traits could have
been reduced in number by excluding those that appeared to
be redundant

(e.g., traits 20 and 70) or by collapsing sev­

eral levels of one trait into a single value
30 to 42, 43 to 49, 78 to 84, etc.).

(e.g., traits

However, such pr o ­

cesses of eliminating and combining would have involved
differentiating the traits according to a priori criteria,
while the objectives of this study required that a priori

36

decisions be avoided as much as possible.

The results of

deciding a priori to combine several levels of a trait could
be invalid in the sense of not reflecting the real inter­
relations among the traits.

This situation was illustrated

by the manner in which traits 43
other traits:

to

49 correlated with

43 belongs to one cluster of traits;

and 47 belong to another;

45, 46,

and 48 and 49 belong to a third

(

cluster.

No reasonable a priori criteria were available

for dividing or combining these levels prior to the process
of correlation; the divisions between traits resulted from
actual relationships inherent in the data.

Of course, all

of the traits used in this study admittedly result from a
priori considerations to the extent that definitions of
terms and procedures are a priori, but that issue is beyond
this study.

The traits were accepted as data in the literal

s ense.
Two different forms were used for 17 of the traits:
one form represented the values of the traits as a percent
of another trait
170, 172, 177,

(traits 163,

178,

179,

180,

164,

165, 166, 167,

168, 169,

181,

182, 183, and 184);

the

other form represented the same 17 traits as whole numbers
(203, 204, 205,

206,

207, 208,

215, 216, 217, 218, and 2 1 9 ) . 1

209,

210,

211, 212,

213, 214,

These traits expressed

t r a i t s 178, 181, and 216 were later dropped from
the analysis because they showed no useful correlations,
and 210 was excluded as duplicating variable 170.

37

departmental expenditures,
characteristics

sources of funds, and faculty

(rank, highest d e g r e e ) .

The set of traits

using percents was chosen for the final analysis, as
explained below.
Various subsets of traits could have provided the
basis for clustering departments,

for example,

only those

traits that pertained to curriculum matters or only those
that pertained to budgetary items.

The present study used

all of the traits currently available in an objective form,
because, since no similar clustering process had been car­
ried out previously, no precedents for selection existed in
the research literature.

Thus, a study using all of the

available objective traits appeared to be a reasonable first
step, to be followed by other clustering procedures based on
subsets of traits.
Further, previous efforts to classify or group
departments

(as discussed in Chapter I, above) have been

based on such concepts as the department's mode of gover­
nance or its value "climate," and have employed interviews
or opinion surveys to obtain the data.

The viewpoint of

this study rests on an assumption that all of the essential
aspects of a department
objectives,

(governance, members,

subject matter,

and functions)

resources,

are equally

important and worthy of serious study, although all aspects
have not actually been given equal treatment in research.
Ironically, research has chiefly treated those aspects

(or

38

those facets of aspects)

that are least amendable to

scientific method, the political and socio-psychological
r e l a tionships.

Methodology
The method used in the present study was a combina­
tion of the techniques of correlation and multiple-groups
analysis embodied in a system of computer programs developed
by Hunter and Cohen

(1969)

3600 Extended F o r t r a n ) .
listed in Appendix C.

called PACKAGE

(written in CDC

The routines from PACKAGE are

Since the purpose of the present

study was to group departments according to empirical traits
relating to departmental functions,

this method was appro­

priate for interrelating the departments in terms of actual
relationships inherent in the data.

The method of this

study, often referred to as a cluster analysis,
of factor analysis.
farious)

is a variety

In terms of this study's data

and objective

(multi­

(grouping e n t i t i e s ) , factor analysis

was the m o s t appropriate and efficient method.
Cattell

(1966)

defined as the general purpose of

factor analysis:
to represent or explain observed covariational
relations among many experimental variables in
terms of linear dependencies on, and relations
among, a much reduced number of "ideal,"
"intervening," or "abstract" conceptual
variables [i.e., factors] (p. 174-5).

39

Factor analysis operates upon data

(which can be

almost any sort of ordinal or dichotomous information)
order to find a meaning in the data's variability.

in

To the

degree in which the data actually possess a p a t t e r n or
uniformity,

factor analysis will uncover it.

One can view

factor analysis as a means of finding "the minimum number
of independent coordinate axes necessary to plot
the variation in vectors in the space"

(reproduce)

(Rumrael, 1970, p. 15),

or in other words, to find the d i m e n s i o n s of the data as
reflected in the coordinate axes of the vector space defined
by the data.

The degree of regularity actually existing in

the data will be reflected in the number and strength of the
dimensions uncovered by a factor analysis.

Factor analysis

can handle a large number of variables simultaneously, and
can handle complex interrelationships as they occur in a
concrete situation .(rather than abstracted in an experi­
mental situation) .

For these reasons,

factor analysis

is especially useful for the multifarious data of social
sciences,

such as the data that describe the functions of

university departments.
Cattell has stated how factor analysis applies to
these data:
Factor analysis . . . i s the ideal method of
open exploration in regions unstructured by
present knowledge. . . . Factor analysis,
carried out on the correlation coefficients,
shows us how some variables can be grouped
together because they behave in the same way,

40

and it proceeds to delineate new independent,
underlying factors which may be responsible for
these g r o u p i n g s . . . . It also groups the vari­
ables together in ways which permit one to
synthesize new entities.
These new entities
are now themselves to be considered as v a r i ­
ables— though far fewer than the initial raw
variables. . . .
There is no point in working out . . . precise
laws about the relations between variables until
we have chosen the s i g n i f i c a n t variables. . . .
The factor analyst is suspicious of choosing the
important variables a p r io ri , no matter how selfevident their significance may seem to the
experimenter. . . . In social phenomena, for
example, the surface variables have been known
to mankind so long that most relationships which
intelligent men could perceive have long been
perceived (Cattell, 1952, pp
14-16, 19).
Factor analysis has advantages where (1) the
number of variables to be watched over and
thought about is bewilderingly large, (2) there
has been little success after several years in
reaching agreement on the major concepts, and
(3) there is good reason to expect complex
interactions, which are not easily experimen­
tally separable by manipulation or control,
notably where most phenomena are multiply
determined (Cattell, 1966, p. 175).
One can easily see how the many disparate data describing
the academic departments of a university fit the category
of variables especially amenable to factor analysis.
Factor analysis can be used for many objectives:
finding patterns of interrelationships among data; achieving
data reduction or parsimony; revealing the structure of a
field of interest;

and grouping entities into a classifica­

tion or an empirical typology

(Rummel,

1970, pp.

29-32) .

The purpose of factor analysis follows from its general
function;

it proceeds by seeking the sources of variability

within the data.

Since the regularity achieved in factor

analysis reflects the clusters of vectors inherent in the
data matrix, the result of a successful factor analysis will
necessarily be the definition of a "new" smaller group of
variables

("factors") which can represent or explain the

larger number of the "given" variables.
Factor analysis is a valuable tool for scientific
research because it especially is concerned with the general
scientific values of p a r s i m o n y and s i m p l i c i t y .

Fruchter

(1954) states:
A basic assumption of factor analysis is that
a battery of interrelated variables has common
factors running through it and that the scores
of an individual can be represented more
economically in terms of these reference
factors (p. 44).
Cattell considers factor analysis "a scientifically p a r s i ­
monious way of accounting for the behavior of many variables
by introducing
(1966, p.

179).

'ideal'

intervening variables or concepts"

Thus factor analysis fulfills an essential

requirement of a scientific procedure because it is e f f i ­
c i e n t ; it produces results more economical than the data to

which it is applied.
Factor analysis also serves another general scien­
tific value by its quality of being h e u r i s t i c or r e v e l a t o r y ,
amenable to further discovery, guidance,
According to Burt:

1

or clarification.

42

The primary aim of factor analysis is exploratory.
It seeks to d i s c o v e r principles of classification;
it is these that constitute what have come to be
called, somewhat inappropriately, "factors" (in
Cattell, 1966, p. 269).
In the present study, the variety of factor analysis
that has been used— usually called a cluster analysis—
involved two basic operations:

(1 ) clustering the variables

by visual inspection of a matrix of correlation and a matrix
of similarity coefficients, which had been rearranged to
present the variables in the order of their correlations and
similarities

(starting from the h i g h e s t ) ; and

(2 ) a multiple-

groups analysis which produced a matrix of the variables'
loadings on the clusters defined in step

(1).

Each of these

parts included several steps as the processes were repeated
in order to achieve the most satisfactory clusters.
Multiple-groups analysis has been described by
Rummel
1.

2.

3.

(1970, pp.

337-338)

as f o l l o w s : 1

[Multiple-group analysis] is a method for
classifying variables into groups or test­
ing for the existence of specified group
factors in a domain.
On the basis of their correlations, the
variables must be clustered into linearly
independent sets.
This requirement makes
the technique most useful when there is
prior knowledge or theory as to what these
groups should be (Fruchter, 1954, p. 87).
The number of factors depends on the rank
of the correlation matrix, i.e., the number
of linearly independent groups of variables.

^ l s o discussed in Harman

(1967), pp.

216-230.

43

4.

Each variable tends toward a complexity of
one, i.e., a loading on only one factor.
5.
The factor complexity is in terms of bipolar
group factors, with a tendency toward these
factors being nonoverlapping.
6 .
The variance contributions of factors are
approximately level.
7. The
procedure is sensitive to the grouping
of the variables, since an u n r o t a t e d solu­
tion may be altered by changing the prior
grouping of variables.
8 .
Because of the possible combinations of
variables and the indeterminateness inherent
in the centroid itself, the multiple-group
solution is not unique for a correlation
matrix.
9. The
computational labor is less than that
for the centroid technique for large
matrices.
10.
The technique extracts all the factors at
once.
11.
The factors may be mutually oblique.
12.
The factors are a linear fit to each of the
linearly independent sets of variables.
The departments could also have been grouped on the
basis of empirical variables by other varieties of factor
analysis, such as the principal axes technique,
by an oblique or orthogonal rotation.

followed

Such an analysis

would produce a matrix displaying the d e p a r t m e n t s ' factor
loadings,

and departments that have similar patterns of

factor loadings could then be defined as a group.

From the

analysis one could also produce estimations of the t r a i t s '
scores for the set of groups.

The results of such p roce­

dures could be hard to interpret.

If each department loaded

highly on several factors, a clear pattern of department
groups might be difficult to determine.
number of traits is large,

Also, since the

it would probably be difficult

44

to interpret the meaning of a group in terms of the traits
that have high or low scores on it, since to characterize
each group, an investigator would have to consider the
scores of 167 traits in order to discover those traits that
were most significant in the interrelations of the de p a r t ­
ments forming each group,
In such a typical factor analysis, characterization
of the groups would f o l l o w the determination of the depart­
ments'

relations to

(loadings on)

the groups.

In the method

used in this study, characterization of the trait-clusters
precedes

the grouping of the departments.

The f a c t o r s in a

factor analysis could be interpreted as p r i n c i p l e s o f i n c l u ­
sion for the groups of departments,

but the factors would

have to be defined by reference to the traits'
each group.

scores on

Listing the departments with high loadings on

a group w ould not be sufficient to define the group.
In the procedures followed in this study, the pr e ­
liminary trait-clusters were formed by careful inspection
of the interrelationships among the traits as shown in the
matrix of correlations.

The investigator could therefore

be aware of the content of the trait-clusters at each stage
and could develop a sense of the nature of each traitcluster as the process continued.

In the second part of

the procedures, when the departments were grouped, the
operation was based on trait-clusters that had already been
characterized.

Since the contents of the trait-clusters

45

were known, difficult situations in the clustering process
(when a department's correlations did not clearly indicate
a group membership)

could be analyzed in terms of the

characters of the trait-clusters involved as well as the
d e p a r tments.
This method facilitated the process of defining the
principles of inclusion

(the departmental "types") because

both the variable-clusters and the departmental groups could
be clearly seen.

A trait-cluster could be seen from the

start as a composite, and the definition of a trait-cluster
could be derived from its components.

Procedures I;
Traits

Clustering the

Step one was a clustering of traits 1-77.

of traits in Appendix B.)

(See list

The entire list of traits could

not be accommodated in one program since a maximum of 150
variables can be processed at one time in the PACKAGE system
on a CDC 3600 computer with a 32K core memory.

Traits 1-77

were used because they appeared to be a reasonable sub-set
of traits.

Step one involved three computer runs, with a

progressive refinement of the clusters after each run.

The

final result was nine trait-clusters and a set of residual
traits.

These nine clusters had high internal consistency;

46

their a l p h a s 1 ranged from 85 to 98, and the average alpha
was 92.3.
Throughout this step and succeeding ones,
criteria defined by Hunter and Browne

(1971, p.

three

3) were used

to form the trait-clusters:
1.

2.

3.

Items in a cluster should be reasonably
well correlated with each other (i.e.,
internally c o n s i s t e n t ) .
The items in a cluster should be reasonably
parallel, i.e., all exhibit the same p a t ­
terns of correlations with other clusters.
They should be reasonably homogeneous in
content.
Internal consistency of trait-clusters and department-

groups was indicated by the correlations of the items within
each cluster or group and by the "alpha" indexes of interr^l
consistency obtained in the multiple-groups analysis.

In

all clustering operations, most of the correlations within
clusters were high, above 65; and almost all were above 50.
Parallelism of items within clusters was indicated by a
matrix of similarity c o e f f i c i e n t s 2 obtained from the SQRR

1 Alpha, also called coefficient alpha, is a measure
of the internal consistency of a cluster or group, based on
the average correlation among the components (Nunnally,
1967) . An alpha of 60 or higher indicates moderate con­
sistency; 80 or higher indicates high consistency.
Alpha
is also an estimate of reliability for a group of variables
forming a scale.

2A similarity coefficient is an index of the degree
to which two variables have a similarity in pattern of
correlations with other variables in a set.
It can be
interpreted approximately the same as a product-moment
correlation, i.e., bounded by plus-one and minus-one.

47

program

(see Appendix C ) .

The matrixes of similarity

coefficients closely reflected the matrixes of correlations.
Homogeneity of clusters was evaluated by reference
to the particular items combined in £ach cluster, at each
stage of the procedure.

However, the criterion of ho m o g e ­

neity was applied sparingly in the process of clustering
traits.

Since homogeneity,

like relevance,

is a concept

essentially defined by the person who uses it, therefore,
it was not entirely appropriate in a study which aims at a
high degree of objectivity, such as the present one.

In

those few instances where the traits being clustered corre­
lated almost equally with traits that appeared to belong to
different clusters,

the nature of the traits as well as

their correlations were considered in forming clusters.
Initial steps in clustering both traits and departments were
made with reference only to the list numbers, or— one could
say— the first stage of clustering was done "blind."
St ep

two was a clustering of traits

78 to 198 which

produced seventeen clusters and a residual set of traits, in
two computer runs.
one.

The procedure was the same as in step

The multiple-groups analysis produced alphas for these

seventeen clusters of 76 to 97, with a mean alpha of 89.5.
In step three traits 16 3 to 184 were replaced by
traits 203 to 219, which represented the whole figures on
which the percentage figures

(163-184) had been based

(as

described previously on page 36).

A second set of clusters

was produced by correlating this second set of traits.
In step f o u r the two sets of clusters based on the
two series of traits

(1 to 77, 78 to 219) were combined.

The final result was two sets of trait-clusters:

a set of

16 clusters using the percentage figures; and a set of 13
clusters using the whole number figures.

The analysis of

two sets of traits was carried out in order to see whether
two different sets of trait-clusters would be produced, as
was indeed the actual result.
The set of trait clusters that included the traits
in percent form was considered more appropriate for the
following reasons:
departments'

this study was intended to focus on the

functions rather than their size alone;

i.e.,

on the traits which represented a department's operations,
not on the correlations of each trait per s e .

When certain

traits were expressed as percents of another trait

(e.g.,

salary expenditures expressed as a percent of total expen­
ditures)

then the percentage trait represented an inter­

relation of several traits.

Percentage figures represented

the traits' relative values, not their absolute values.
Also, using percents for certain traits reduced the effect
of size upon the clustering process.

The reason for this

effect is that in large departments, salary expenditures
would in most cases also be large; the salary expenditures
of all large departments w ould correlate highly,

regardless

49

of the relation of salary expenditures to other expenditures
within each department.

Using percentage figures stresses

the relation of variables to other variables within the same
department.

Procedures II:
Departments

Grouping the

In step five,

the same methods were used as in

previous steps, except that the items being clustered were
the 79 departments and the clustering was based on corre­
lations of the departments'

scores on the trait-clusters

defined in steps one through four.
A department's s c or e for a trait-cluster could be
produced in two ways:
within each cluster;

by summing the values of the traits
or by averaging the same v a l u e s . 1

Both

procedures were used in order to see w h a t differences would
em e r g e .
Thus,

four sets of department groups were initially

produced by using the two sets of trait-clusters
above)

(described

and the two types of cluster scores as bases for

grouping d e p a r t m e n t s :
Set A:
Set B:

trait-clusters with percents,
from averaging;
trait-clusters with percents,
from summing;

scores
scores

A v e r a g i n g the values within a trait-cluster did not
function as a linear transformation of the sums of the same
values because the trait-clusters had differing totals of
component traits.

50

Set C:
Set D:

trait-clusters with whole figures, scores
from averaging;
trait-clusters with whole figures, scores
from summing.

One might expect that traits expressed as percents
would exhibit less variance than the same traits expressed
as whole numbers;

also, that clusters whose total values

were obtained by averaging the component traits'

values

would exhibit less variance than clusters whose total values
were obtained by summing the component values.

Therefore,

it was predicted that the four sets of department groups
would exhibit different amounts of variance in the traitcluster scores;

that Set D w ould exhibit the most variance

and Set A the least.

The amount of variance is relevant

because it affects the level of correlations.
The prediction was essentially correct.

Use of

summing or averaging produced marked differences in the
amount of variance within each department group.

However,

the use of the sets of traits with percentages or whole
figures for certain traits produced relatively little d i f ­
ference in the amount of variance within each department
group.
Regardless of the differences in variance, each
set of department groups was essentially distinct.
between the four sets was small.

Overlap

Only five pairs of depart­

ments and two triples appeared in all four sets, after the
first multiple-groups analysis.

51

On the basis of the clarity of pattern produced by
the multiple-groups analysis,

Set A

(percents, averaging)

appeared to be the most ef f i c i e n t of the four sets for the
purposes of this study.

This conclusion is consistent with

the theoretical expectation that the set of groups in which
the variance between departments is smallest will produce
the smallest number of high correlations between groups and
the most clear-cut overall patterns of correlations . 1

1 In set A
(percents, averaging) correlations were
appropriately high within groups (although lower overall
than in Set B) but departments showed few high correlations
outside their own groups.
The groups correlated highly with
only a few other groups.
Thus, the result was a relatively
clear pattern, with high within-groups correlations and low
between-groups correlations.
Set B (percents, summing) contained a large number
of high correlations ( r = 7 0 + ) .
Many departments correlated
highly with several g r o u p s , and many groups correlated
highly with several other g r o u p s , but showed no clear p a t ­
tern of correlation.
Some groups that correlated highly
with the same group did not correlate highly with each other,
and so on.
The overall pattern was confused.
The multiple-groups analysis of Set D (whole figures,
summed) shows a complex pattern in which most of the groups
correlated highly with several other groups in an overlap­
ping pattern that did not facilitate reduction in the number
of groups.
Set D resembled Set B in its high between-group
correlations.
Thus, the process of summing the values
within the variable-clusters was apparently not efficient
for this study.
Using the cluster scores obtained by sum­
ming the variables' scores as the basis for correlating the
departments increased the variance between departments and
thereby increased the correlations between departments.
Averaging the variables within a cluster produced less
between-department variance and lower between-group
correlations.
In the multiple-groups analysis of Set C (whole
figures, averaging), most variables correlated highly with
variables in the same groups but not with variables in other
groups.
In this aspect, Set C resembled Set A (percents,
averaging) but not sets B and D.
The pattern of variable

52

The choice of Set A as most suitable for further
analysis was determined by the following criteria:
(1 ) c la ri ty of the pattern of correlations among the
departmental g r o u p s ;
(2 ) e f f i c i e n c y in the reduction of the original set
to a smaller number of groups;
(3) c o n s i s t e n c y with the overall goals and expectations
of the study, i.e., with the goal of developing a
method suitable for classifying a collection of
departments varying greatly in size as well as
type.

Summary
The primary objective of this study was to group
university departments on the basis of objective variables
describing departmental functions, by means of a process
employing the techniques of correlation and multiple-groups
analysis.

The second objective was to reveal the dimensions

or clusters of the variables describing departmental func­
tions, using the same process.
The population of interest included 79 of the aca­
demic departments at Michigan State University.

The de p a r t ­

ments are listed in Appendix A.
The data included 16 7 departmental traits gathered
chiefly from reports routinely prepared at Michigan State
University.

The traits are listed in Appendix B.

correlations was relatively clear-cut in Set C.
However,
the correlation of groups in Set C was not efficient in the
sense that it did not facilitate reduction of the number of
groups.
In Set C, as in sets B and D, the correlations pr o ­
duced an overlapping pattern of department groups.

CHAPTER IV
THE RESULTS

The Clusters of Traits
As a result of the clustering operations and the
multiple-groups analyses,

the 167 departmental traits were

amalgamated into sixteen trait-clusters.

The full list of

the traits that comprise each cluster are exhibited in
Appendix D.

The correlations and alphas of the trait-

clusters are shown in Table 1 and Figure 1.

These trait-

clusters could be considered both as dimensions inherent
in the traits and as categories into which the traits can
be d i v i d e d . 1
The trait-clusters have been described in terms of
their salient components.

To determine the traits that are

salient within a cluster,

reference was made to each trait's

^rhe trait-clusters could also be termed "factors,"
since they resulted from a form of factor analysis.
The
terms "cluster" and "group" have been used in place of
"factor" in order to clarify the relations among the
elements of the present study.
Confusion could arise
because both trait-clusters and department groups can
be considered factors.

53

54

Table 1.

Cluster

Alphas for Trait-Clusters

Number of Traits

Alphas

C- 1

53

98

C- 2

10

89

C- 3

25

97

C- 4

18

94

C- 5

3

78

C- 6

4

92

C- 7

4

87

C- 8

3

69

C- 9

7

93

C-10

10

93

C-ll

9

89

C-12

4

71

C-13

8

88

C-14

6

97

C-15

1

100

C-16

2

81

55

1
2

56

3

48

4

46

5

31

6

51

•24 -35

44

-35 -31

52

7

28 -39

8

-39 -33

-1

■31

19

32

10

10

14

24

30

11

24

-7

■15

12

32

34

■14

13

13

36

19

14

-12

15

13

51

-8 -16 -10

-4

32

14

35

■13 -30

-3

-7

•28

27

-4

-5

-6

13

16

•10 -11
23

36

-21
■22

Figure 1.

-2

-7 -10 -21 -15

•24 •21

-15 -20

48
41

•30 -14

9

16

33

-4

■22

•21

26

25

33

-9

•10

-4

36

-15 -13
24

-7

-3

■14

9

10

11

-4
•22 -13
13

-4

•11

10

-5

14

15

12

Correlation of Trait-Clusters.

13

16

56

l o a d i n g 1 on its cluster as determined by the multiple-groups

analysis.

Traits with loadings of 67 or above were consid­

ered salient, those with loadings of
considered d o m i n a n t . 2

86

or above were

(See list in Appendix D for salient

and dominant t r a i t s .)
The trait clusters have also been described in terms
of the departments that had high or low scores on each traitcluster . 3

The eight departments

(10 percent of the total

population) with highest and with lowest scores were listed
for most of the trait-clusters.

This form of characteriza­

tion was also intended to facilitate relating the traitclusters to the concrete situation within the university.
The summary descriptions of the sixteen traitclusters are as follows:

*A loading is a correlation coefficient between
traits and trait-clusters (and also between departments and
department g r o u p s ) . It indicates the amount of variation a
trait has in common with a trait-cluster.
The square of a
loading times 1 0 0 equals the percent of a trait's variance
accounted for by a trait-cluster; e.g., a loading of 67
indicates that 45 percent of a trait's variance is in common
with a certain trait-cluster, and a loading of 8 6 indicates
74 percent of variation in common.
2Selection of these two points was essentially
arbitrary, although guided by certain considerations, such
as the percentage of common variance indicated by loadings
of 67 and 8 6 (as mentioned in footnote 1).
Common variance
less than 45 percent or thereabouts would hardly justify
calling a trait "salient" within a certain trait-cluster.
Also, dividing the loadings at 67 and 8 6 matched the gaps
in the distribution of loadings.
3The scores represent the average of the values for
' all the traits in a cluster, for each department.

57

C l ust er 1
(52 traits)

Graduate and upper division curriculum and
enrollments
Productivity:
degrees granted; student
credit hours at graduate and upper
division levels
Graduate assistants; large B - f a c u i t y 1
Student majors
General Fund expenditures
Emphasis on independent study, variable
credit (graduate)
Teaching loads:
full-time faculty,
chiefly in independent study, variable
credit; graduate assistants, low load
in classes
High s c o r e s : 2
Labor and Industrial Relations
Crop and Soil Science
Audiology and Speech Science
Food Science
Medicine
Pathology
Religion
Forestry
Low s c o r e s : 3
Dairy
Agricultural Engineering
Large Animal Surgery and Medicine
Romance Languages
Philosophy
History
German and Russian
Park and Recreation Resources

^ - f a c u l t y includes academic staff members not
classed as instructors, assistant professors, associate
professors, or professors.
2In order of size of scores:
the first department
listed had the highest score, the last one had the lowest
score.
3A 1 1

departments listed had the same low score.

58

Cluster 2
(10 traits)

Undergraduate courses (upper and lower
division)
Students seeking teacher certification
Small recitation sections (11-20 students)
High s c o r e s :
Psychology
Microbiology and Public Health
Biochemistry
Physics
Crop and Soil Science
Agricultural Economics
Chemistry
Mathematics
Low s c o r e s ;
Medical Technology
Chemical Engineering
Accounting and Financial Administration
Packaging
Advertising
Business Law and Office Administration
Religion
Astronomy

C lu s te r 3
(24 traits)

Lower division curriculum and enrollments
Productivity:
student credit hours at
lower division level
Full-time faculty (A-facuity ) 1
Recitation method
Teaching load:
full-time faculty,
classes, medium to high
High s c o r e s :
Psychology
Mathematics
Chemistry
Sociology
English
History
Political Science
Economics

^ - f a c u l t y includes assistant professors,
professors and professors.

associate

59

Low s c o r e s ;
Park and Recreation Resources
Poultry Science
Medicine
Astronomy
Religion
Psychiatry
Human Development
Medical Technology

C l us te r 4
(18 traits)

Part-time faculty
Total expenditures
Teaching load:
part-t.'me faculty in
independent study, variable credit,
low load
High s c o r e s :
Humanities
Mathematics
American Thought and Language
Social Science
Natural Science
Romance Language
English
Psychology
Low s c o r e s :
Animal Husbandry
Biophysics
Astronomy
Psychiatry
Pharmacology
Park and Recreation Resources
Medical Technology
Poultry Science

C l u st e r 5
(3 traits)

Experiment Station and Extension Service
support
Labor expenditures (percent of total
e x p e n ditures)
(Although these traits correlated highly,
departments' scores for this traitcluster were not m e a n i n g f u l .)

60

Cluster 6
(4 traits)

Sponsored research, expenditures (percent
of total expenditures)
Supplies, service, equipment expenditures
(percent of total expenditures)
High s c o r e s :
Small Animal Surgery and Medicine
Large Animal Surgery and Medicine
Anatomy
Microbiology
Medicine
Pathology
Pharmacology
Physiology
Low s c o r e s :
(The following departments all
had the same score.)
Religion
Business Law and-Office Administration
Astronomy
American Thought and Language
Nursing
Humanities
Natural Science
Social Science
Medical Technology

C lu st er 7
(4 traits)

Faculty time:

research

High s c o r e s :
Natural science
Chemistry
Geography
Music
Art
Geology
Physics
Agricultural Engineering

61

Low s c o r e s :
(The following departments all
had the same score.)
Anatomy
Astronomy
Biophysics
Communication
Economics
Human Development
Labor and Industrial Relations
Large Animal Surgery and Medicine
Medical Technology
Pharmacology
Poultry Science
Psychiatry
Religion
Resource Development
Small Animal Surgery and Medicine

C lust er 8
(3 traits)

Faculty time:
administration
Faculty with professional degree
of faculty)

(percent

High s c o r e s :
Pharmacology
Biophysics
Labor and Industrial Relations
Theatre
Psychiatry
Poultry Science
Human Development
Medicine
Low s c o r e s :
English
Romance Languages
Horticulture
Animal Husbandry
Fisheries and Wildlife
Resource Development
Music
Agricultural Engineering

62

Cluster1 9
(7 traits)

Emphasis on independent study and variable
credit courses (lower division)
Graduate assistants, wit h low load in
independent study, .variable credit
High s c o r e s ;
History
Romance Languages
English
Music
Mathematics
German and Russian
Art
Linguistics
Low s c o r e s ;
Medical Technology
Biophysics
Pharmacology
Large Animal Surgery and Medicine
Small Animal Surgery and Medicine
Labor .and Industrial Relations
Psychiatry
Human Development

C lu s te r 10
(10 traits)

L a b o r a t o r y 1 method, undergraduate
Lecture sections, undergraduate, large
size (50-200 students)
High s c o r e s :
Music
Political Science
Communication
Criminal Justice
Linguistics
Computer Science
History
Animal Husbandry

l a b o r a t o r y section category is defined by mode of
teaching and .includes.teaching that does not take place in
an actual scientific laboratory.

63

Low s c o r e s ;
(59 departments had the same
score.
The following were randomly
selected from that group.)
Agricultural Economics
Poultry Science
Management
Mechanical Engineering
Astronomy
Nursing
American Thought and Language
Pharmacology

Clu st er 11
(9 traits)

Maleness (faculty, students)
Faculty status (tenure, rank)
High s c o r e s :
English
Agricultural Engineering
Crop and Soil Science
Food Science
Horticulture
Mathematics
Animal Husbandry
Mechanical Engineering
Low s c o r e s :
(36 departments had the same
score.
The following were randomly
selected from that group.)
Philosophy
Management
Computer Science
Human Development
Chemistry
Criminal Justice
Humanities
Medical Technology

64

Cluster 12
(4 traits)

New and transfer undergraduate majors
High s c o r e s :
Marketing and Transportation Adminis­
tration
Religion
Chemical Engineering
Metalurgy, Mechanics, and Materials
Science
Geology
Management
Civil and Sanitary Engineering
Animal Husbandry
Low s c o r e s ;
Social Work
American Thought and Language
Medical Technology
Human Nutrition
Business Law and Office Administration
Family and Child Science
Human Environment and Design
Nursing

C lu s te r IS
(8 traits)

Su b - c o l l e g e 1 curriculum and enrollments
High s c o r e s :
Psychiatry
Medicine
Human Development
Astronomy
Pathology
Linguistics
Anatomy
Dairy
Low s c o r e s ;
Mathematics
Food Science
Metalurgy, Mechanics, and Materials
Science
Economics

1 Sub-college category includes both college prepara­
tory (remedial) courses and short-rterm vocational-technical
courses.

65

Zoology
American Thought and Language
Biochemistry
Marketing .and Transportation A d m i n ­
istration

C lu s t e r 14
(6 t r a i t s )

Graduate-professional curriculum and
enrollments
High s c o r e s :
Poultry Science
Animal Husbandry
Dairy
Crop and Soil Science
Resource Development
Horticulture
Agricultural Engineering
Agricultural Economics
Low s c o r e s :
(20 departments had the same
score.
The following were randomly
selected from that group.)
German and Russian
Linguistics— Oriental and African
Languages
Romance Languages
Management
Advertising
Theatre
Biophysics
Nursing

C l u s t e r IS
(1 trait)

Educational Development Program
of total expenditures)

(percent

High s c o r e s ;
Human Development
Pharmacology
Physics
Chemistry
Biophysics
Family and Child Science
Medicine
Large Animal Surgery and Medicine

Low s c o r e s ;
Theatre
English
Humanities
American Thought and Language
German and Russian
Romance Languages
History
Religion

C lu st er 16
(2 traits)

Faculty time:

service

High s c o r e s :
TV and Radio
*Music
Business Law and Office Administration
Journalism
Mechanical Engineering
Zoology
Political Science
Psychology
Social Work
Urban Planning and Landscape
Anatomy
Low scores:

(All other d e p a r t m e n t s .)

*The following ten departments all have the same
score.

67

The Groups of Departments
The 79 departments were formed into preliminary
groups on the basis of their correlations across the six­
teen trait-clusters; then a multiple-groups analysis was
performed on those preliminary groups.
procedures,

As a result of these

the 79 departments were amalgamated into eleven

departmental groups.

The internal consistency of the groups

is indicated by their alpha indexes as shown in Table 2.
Their correlations are shown in Figure 2.

Table 2.

Group

Department-Group Alphas

Number of Departments

Alpha

1

15

95

2

12

94

3

6

87

4

9

92

5

5

89

6

6

89

7

4

88

8

7

94

9

6

85

10

5

86

11

4

73

68

-7

10
11

-7

30

51

25

-4

-53

-27

-59

-32

-20

-53

52

33

-20

-22

-21

-21

-8

-19

-33

-6

44

-7

-29

12

25

14

-16

34

36

-3

40

-6

16

-32

33

-12

-12

-52

-35

-39

-21

20

-29

10

10

Figure 2.

Department-Groups Correlations

11

69

The department groups were as follows:
GROUP 1:

Agricultural Sciences

♦1
2

♦♦3
♦4
♦♦5
6

♦7
♦8

*
11
♦12

♦13
49
♦51
♦♦53
GROUP 2:
♦21

♦23
25
♦26
♦♦27
♦♦28
♦35
♦♦36
38
♦40
54
♦♦58
GROUP 3:
♦52
♦55
♦57
60
♦65
♦67

Agricultural Economics
Agricultural E n g i n e e r i n g ^ ^
Animal Husbandry
Crop and Soil Science
Dairy
Fisheries
food Science
Forestry
Horticulture
Hark and Recreation R e s o u r c e s ^ ^
Poultry Science
Resources Development
Biochemistry^^
Botany and Plant Physiology
Entomology
Business and Engineering
Religion
Accounting and Financial Administration
Economics^^
Hotel, Restaurant, and Institutional M a nagement♦ ilk *
Management
Marketing and Transportation Administration
Chemical Engineering
Civil and Sanitary Engineering
Electrical Engineering and Systems Science
Metalurgy, Mechanics, and Materials Science
Geology
Statistics
Scientific Disciplines
Chemistry
Mathematics
Physics
Anthropology
Psychology
Sociology

♦Loading of 67 to 85 on departmental group.
footnote 1, page 56.
♦♦Loading of

86

See

or above on departmental group.

♦♦♦Refer to notes at end of list.

70

GROUP 4:
**14
15
**16
**17
*18
*20
**22

44
62
GROUP 5:
**45

46
*47
*50
*78
GROUP

6

:

*73
**74
*76
*77
79
**80
GROUP 7:
**69
**70
71
**72
GROUP

8

:

*24
**32
**33
*39
59
*66
**68

Humanities
Art
English***
German and Russian
History
Linguistics***
Philosophy
Romance Languages
Human Environment and Design***
Geography
Health Sciences I
Human Development
Medicine***
Psychiatry
Biophysics
Pharmacology
Health Sciences II
Anatomy
Large Animal Surgery and Medicine
Microbiology and Public Health
Pathology
Physiology***
Small Animal Surgery and Medicine
University College
American Thought and Language
Humanities
Natural Science***
Social Science
Service Oriented
Business Law and Office Management
Journalism
TV and Radio
Mechanical Engineering
Zoology
Social Work
Urban Planning and Landscape

71

GROUP 9:

Mixed

*19
29
31
*37
**61
**64
GROUP 10:
34
*43
*48
56
**75
GROUP 11:
10

**30
42
63

Music
Advertising***
Communications * * *
Computer Science
Criminal. Justice
Political Science
Mixed
Theatre
Human Nutrition and Foods***
Astronomy***
Nursing
Medical Technology
Mixed
Packaging***
Audiology and Speech Science**
Family and Child Science***
Labor and Industrial Relations

Not es

on D e p a r t m e n t a l

indicate departments;

Groups:

Numbers prefixed "D-"

numbers prefixed "G-" refer to

department groups.

Group 1
D-2

correlates .64 wit h D-15, but their correlations are
not parallel.
D-2 correlates with D-4 (.63) and D-9
(.6 6 ) but loads only .45 on Group 1.

D— 11

correlates .76 with D-43, but loads only .53 on
Group 6 to which D-43 belongs.
D-ll loads .62 with
its own group, G-l.

D-49

loads only .58 on Group 1, but fits better there
than in any other group.
D-49 correlates .75 with
D - 5 7 , but not with D-57's group, G-3.
D-49 is
tangential to G-l and G-3.

72

Group 2
D-25

loads .70 on Group 3, but shows strong correlations
with components of its own group, G-2 (loads .61).
D-25 correlates .77 with D-55 in G-3.
D-25 is
tangential to G-2 and G-3.

D-26

correlates .56 wit h Group 9 and .69 wit h its own
group, G-2.
D-26 correlates .67 with D-61 in G-9.

Group 3
D-55

correlates

.77 with D-25 in Group 2.

Group 4
D-15

correlates .64 with D-2 in Group 1; correlates
with D-14; and loads .56 on Group 4.

D-18

.61

correlates .69 with D-29 in Group 9.

D-44

correlates .73 with D-56 in Group 10.
D-44 loads
approximately the same on both G-4 (.50+) and G-10
(.50-), but fits better in G-4.

Group 5
D-46

Group

correlates .75 with D-48 in Group 10.
tangential to G-5 (load .51) and G-10

D-46 is
(load .44).

6

D-79

is tangential to G - 6

(load .61) and G-5

(load .54).

Group 7
D-71

loads only .36 with Group 7, its highest group
correlation.
D-71 correlates .72 with D-62 in G-4.

Group 9
D-29

correlates

.69 with D-18 in Group 4.

D-31

loads .42 on Group 9 and .38 on Group 5;correlates
about .50 with D-50, D-60, D-64, and D-78.
No other
high correlations.

D-61

correlates

.67 with D-26 in Group 2.

D-43

correlates

.76 with D-ll in Group 1.

D-48

correlates

.75 with D-46 in Group 5.

Group 10

73

Group 11
D-30
&
D-42

correlate .75.
D-42 correlates weakly w ith D-63
and D-10.
D-30 correlates weakly wit h D-63 (.59)
and D-10 (.40).
D-63 and D-10 correlate at only
.37.
Group 11 is integrated chiefly by the
correlations of D-3 with the other three, which
are not intercorrelated.

Descriptions of Departmental Groups
Descriptions of the departmental groups were
developed from the groups'
clusters.

scores on each of the 16 trait-

These group scores are formed by averaging the

scores of the departments within a group.

(The department

scores are the averages of each individual department's
values for the traits within a t r a i t - c l u s t e r .)
were expressed as standard scores
deviation of 1).

(mean of

Department groups'

0

The scores

, standard

scores are shown in

Table 3.
The group scores were divided arbitrarily into five
categories of approximately equal size:
medium, medium-low, and l o w .

high, medium-high,

1

*The distribution curve of department group scores
was skewed to the right because of eleven group scores of
1.50 (standard score) or above.
Therefore, the median was
lower than the mean, and the scores in the "medium" category
were all below the mean of zero.
The scale of scores used
to separate the categories is as follows:
High
Medium-High
Medium
Medium-Low
Low

.30
-.15
-.30
-.45
-1.50

and abo
to
.29
to -.16
to -.31
to -.46

74

Table 3.

Department. Groups:

Scores on Trait-Clustersa

Department Groups
Trait
Cluster

1

1

.15

-.17

-.64

-.73

.21

-.02

-.76

-.14

2

.55

-.44

1.17

-.29

-.27

.33

-.32

3

-.32

.11

1.59

.35

-.60

-.20

4

-.44

-.12

.85

.46

-.51

-.40

5

-.04

-.02

.37

.32

-.57

6

-.19

-.27

-.26

-.26

.41

2.63

7

-.13

-.17

.31

.31

-.42

8

-.07

-.45

-.22

-.57

9

-.32

-.11

.39

10

-.16

-.17

11

.57

12

2

3

!L

5

6

7

10

11

-.33

-.45

4.35

-.36

-.23

-.44

-.13

-.54

-.06

.17

-.49

-.27

2.24

-.24

.03

-.41

-.40

-.51 -1.43

.22

.32

1.06

-.34

-.39

-.32

-.31

-.35

-.10

-.25

1.27

.03

-.07

-.21

-.26

2.05

.36

-.28

-.34

-.05

.71

.16

1.50

-.69

-.57

-.28

-.13

.27

-.32

-.23

-. 14

-.06

-.22

-.15

-.22

-.13

1.93

-.22

-.22

-.23

-.01

.30

-.33

-.31

-.02

-.14

-.27

-.25

-.25

.58

.77

.38

-.48

-.46

-.17

-.62

-.25

.02 -1.15

-.68

13

-.12

-.41

-.37

-.18

2.03

.42

-.20

-.14

-.02

.20

-.33

14

1.50

-.46

-.29

-.48

-.47

.05

-.45

-.47

-.36

-.17

-.20

15

.40

-.57

.80

-.93

1.65

.63

-.94

-.23

-.51

-.25

.55

16

-.36

-.36

.01

-.36

-.36

.01

-.36

2.43

-.37

-.36

-.36

aStandard scores.

4

8

9

75

The general categories of traits that were used to
describe the groups are as f o l l o w s ;
C u r ri c ul u m:

type of teaching

faculty, full-time and part-time;
assistants

Staff:

Faculty

time:

graduate

research, administration, service

degrees; student credit hours,
graduate and upper division and lower division

Productivity:

o f funds:
sponsored research, Experiment
Station and/or Extension; Educational
Development Program

Sources

total; percent from General Fund;
percent for supplies and services; percent
for labor

Expenditures:

F ac u l t y

s ta t us a n d gender.

In the descriptions of the departmental groups, only
the traits for which a group had high, medium high, and low
scores have been included.

Traits that have been listed

should be interpreted as "high" unless otherwise indicated.
Traits that have not been listed should be interpreted as
falling within the medium ranges of scores, but not as
irrelevant or missing.
described as high

For example, when a group was

(or low) on graduate curriculum but

undergraduate curriculum was not mentioned, one should
assume that the group had a me d i u m

(or average)

emphasis

on undergraduate curriculum.
The descriptions of the departmental groups are
as f o l l o w s :

76

Group 1:

Agricultural Sciences

(15 departments)

undergraduate courses; graduateprofessional; graduate and upper division (mh )
sub-college (mh); independent study, variable
credit (mh); small recitation sections

C u r ri c ul u m:

many graduate assistants

S t a f f size:

1

(mh)

load:
faculty— chiefly in independent
study, variable credit; graduate assistants,
in classes

T ea c h i n g

F a c ul t y

time:

research

(mh); administration

(mh)

Experiment Station and Extension
Educational Development Program

S o u rc e s o f f und s:

Service;

Entomology, Horticulture, Animal
Husbandry, Dairy

Examples:2

Group 2;

Business and Engineering

(12 departments)

lower division (mh); independent study
variable credit (mh); recitation sections (mh)

Curriculum:

S t a f f size:

faculty

full-time faculty
(mh)

(mh); part-time

Management, Statistics, Marketing and
Transportation Administration, Civil and
Sanitary Engineering

Examples:

Medium-high.
d e p a r t m e n t s as of 1969-1971.
Most of the depart­
ments suggested as examples had loadings on their groups of
90 or more.
These examples have been presented to help
characterize the departmental groups; they were selected
because of their manner of functioning (as indicated by the
analysis of this s t u d y ) . The examples were not intended to
indicate a necessary relation between a departmental group
and a discipline.
This study was not concerned with such a
relationship.

Group 3;

Scientific Disciplines

(6

departments)

lower division; undergraduate courses;
independent study, variable credit (lower div i ­
sion); low emphasis on graduate curriculum;
recitation sections; laboratory teaching (mh);
large undergraduate lectures (mh); many students
seeking teaching certification.

C urr i cu lu m :

large full-time and part-time faculty;
few graduate assistants

S t a f f size:

load:
full-time faculty in classes,
medium to heavy load; part-time and graduate
assistants in independent study, variable credit

Teaching

F a cul ty

high in research;

time:

(service, high for

Psychology)
S ou r c e s o f fu nd s:

Educational Development Program

Chemistry, Mathematics, Physics,
ogy, Sociology

Examples:

Group 4:

Humanities

Psychol­

(9 departments)

lower division; low emphasis on graduate
curriculum; laboratory teaching (mh); independent
study, variable credit; recitation teaching;
large lectures (mh)

C u r ri c ul u m:

large full-time and part-time faculty;
below average in graduate assistants

S t a f f size:

load:
faculty in classes, medium to heavy
load; graduate assistants in independent study,
variable credit

T e a c hi n g

time:
above average in research;
administration

F a cu lt y

Art, German and Russian,
Romance Languages

E xa mp l es :

low in

History,

I
IP
gj's

Group 5;

Health Sciences I

(5 departments)

sub-college; graduate and upper
division (mh); (low emphasis on lower division
curriculum); independent study, variable credit
(graduate)

C u r ri c u l u m :

!

small full-time and part-time faculty;
graduate assistants (mh)

S t a f f size:

F a cu l ty

time:

high in administration

high in sponsored research;
Educational Development Program

S o u r c e s o f funds:

Examples:

Human Development, Psychiatry, Pharma­

cology

Group

6

:

Health Services II

(6

departments)

sub-college; undergraduate courses;
graduate and upper division (mh); graduateprofessional (mh); laboratory teaching,
undergraduate (mh); independent study, variable
credit (graduate); small recitation sections;
large undergraduate lectures (mh); teacher
certification

C u rr i cu l um :

S t a f f size:
F ac u l t y

time:

graduate assistants

(mh)

high in administration;

service

(mh)

high in sponsored research;
Educational Development Program

S o u rc e s o f funds:

Large Animal Surgery and Medicine,
Small Animal Surgery and Medicine

Examples:

Group 7;

University College

(4 departments)

low emphasis on both graduate and
undergraduate curriculum; low emphasis on
independent study, variable credit

Curriculum:

large part-time faculty; small full­
time faculty; few graduate assistants

S t a f f size:

79

time:
high in research
Science Department)

Faculty

University College departments

Examples:

Group

8

;

(chiefly in Natural

Service Oriented

(7 departments)

graduate and upper division (mh); lower
division (mh); sub-college (mh); laboratory
teaching (mh); independent study, variable
credit (mh); recitation sections (mh); large
undergraduate lectures (mh)

C u r ri c ul u m:

full-time faculty
assistants (mh)

(mh); graduate

S t a f f size:

Faculty

time:

high in service;

research

(mh)

load:
full-time faculty chiefly in
independent study, variable credit (mh);
graduate assistants, in class (mh)

Teaching

Group 9:

Mixed

(6

departments)

lower division (mh); sub-college (mh);
laboratory teaching; large undergraduate
lectures; independent study, variable credit
(mh); recitation sections (mh)

C ur ri c ul u m:

faculty full-time

S t a f f size:

(mh) and part-time

(mh)
F a cu lt y

time:

administration

(mh)

load:
faculty in class; graduate assis­
tants in independent study, variable credit

Teaching

Criminal Justice, Political Science

Examples:

Group 10;

Mixed

(5 departments)

sub-college (mh); low emphasis on
lower division curriculum

C u rri cu lu m:

S t a f f size:

small full-time faculty

80

Faculty

time:

high in administration

Medical Technology, Astronomy, Human
Nutrition and Foods

Examples:

Group 11;

Mixed

(4 departments)

graduate and upper division; u nder­
graduate courses (mh); small recitation sections;
independent study, variable credit (graduate);
laboratory teaching, undergraduate (mh); large
undergraduate lectures (mh); teacher certifi­
cation

Cu rr ic u lu m:

S t a f f size:
Faculty

time:

graduate assistants;
administration

large B-faculty

(mh)

load:
full-time faculty chiefly in
independent study, variable credit; graduate
assistants, class load

Teaching

of funds:
sponsored research
Educational Development Program

S o u rc e s

E xam p le :

(mh);

Audiology and Speech Science

Summary
The 167 departmental traits were correlated across
the 79 departments and ultimately clustered into 16 traitclusters.

Then, the 79 departments were correlated across

the 16 trait-clusters and grouped into eleven departmental
groups.

The contents of the trait-clusters are listed in

Appendix D.

The list of department groups is in the chapter,

above.
Each trait-cluster was characterized in terms of
those component traits that had high "loadings" on a traitcluster, as indicated by the multiple-groups analysis.

81

Each department group had a score for each of the
trait-clusters.

Descriptions of each department group were

drawn on the basis of the trait-clusters on which a depart­
m en t group had high or low s c o r e s .

CHAPTER V
DISCUSSION

Problems from the Methodology
The procedures used in this study— a clustering
operation preceding a multiple-groups analysis— were useful
insofar as they permitted the investigator to reduce a large
number of variables
meta-variables

(the 167 traits)

to a smaller number of

(the 16 trai t - c l u s t e r s ) .

This reduction

permitted the investigator to interpret the relations of
traits and departments w ith greater convenience than would
have been possible with the full number of tracts;

and also

satisfied the general scientific values of parsimony and
simplicity.

However,

the procedures had several drawbacks.

For one, the department groups that were developed in the
second phase of the study were determined not by the depart­
ments'

correlations across the 167 traits, but across the

sixteen trait-clusters.

Because of the clustering,

the

particular traits could not function independently to
determine the correlations among the departments.
words,

the particular traits were "lost" or embedded within

their clusters.
traits,

In other

Since some clusters included more component

some traits were more "lost" than others.

82

83

However, particular traits could be related to their
trait-clusters by means of the lo adings produced by the
multiple-groups analysis.

Through the loadings, one could

determine which traits were most salient or "important" for
a cluster

(that is, most completely involved in a cluster);

one could then i n d i r e c t l y relate the traits to the depart*ment groups by referring to a group's scores on the traitclusters

(see Table 3) and observing which traits were

salient within the trait-clusters.

This indirect process

of relating traits to department groups was used to develop
the summary descriptions of department groups, page 73,
above.
In two instances, embedded traits were a serious
problem for interpreting the department groups, as illus­
trated by an apparent contradiction in the description of
Groups 3 (Scientific Disciplines)
For both groups,

and 4 (Humanities ) . 1

their scores on trait-clusters seemed

to indicate that they had hi gh total expenditures and low
General Fund expenditures.

These findings contradicted the

facts that Group 3 actually was high on both expenditures,
and Group 4 actually was a v er a ge on both.

The probable

cause of this situation was that the traits were embedded:
total expenditures in Trait-cluster 4 and General Fund
expenditures in Trait-cluster 1.

Both Groups 3 and 4 had

‘See also Types D and E, pp.

9 8-99, below.

84

low scores on Trait-cluster 1 (chiefly including traits for
graduate and upper division instruction)

and thus their

scores for General Fund expenditures also appeared low, but
incorrectly so.

Similarly, both Groups 3 and 4 had high

scores for Trait-cluster 4 (because they both had large
part-time faculties)

and thus their scores for total

expenditures also appeared high, which was actually correct
for Group 3 but not for Group 4.
The embedded traits appeared to create the greatest
difficulties with respect to the traits for expenditures and
sources of f u n d s .
Another methodological problem arose from the fact
that each trait-cluster carried the same weight in the
process of clustering departments,

although they actually

represented greatly different segments of the departments'
traits.

Some clusters included many traits

(e.g., Trait-

cluster 1, w ith 53 traits) while others contained only one
or a few; also,

some of the large clusters were relatively

heterogeneous in the sense that they incorporated traits
from more than one aspect of university operations
Trait-cluster 1).

(e.g.,

As a result, undue importance was given

to a cluster such as 15, which included only one trait,
Education Development Project funding.
A further difficulty in interpreting the results
of this study arises from the relation of a particular
department to its department group.

The problem is that

85

the specific entities do not always match the general
description.

Departments that have high loadings on their

group may have low scores on one or several of the traitclusters on which the group scored high.

For example,

in

Group 1, the Department of Horticulture loaded .90 on the
g r o u p , 1 but it scored above average on two trait-clusters
and below average on one.

A similar difficulty arose from

the extreme scores of some departments on certain traitclusters, for example,
above

the Department of Music, which loaded

.67 on Group 9, but had an extremely high score for

laboratory sections
sections)

(the category for instrumental practice

and thereby over-emphasized that trait-cluster for

the entire group.

This problem is always inherent in the

relations of the general and the specific, and is therefore
unavoidable.

It is a problem in the interpretation of

r e s u l t s , rather than in the methodology that produced them.
Nevertheless, it should be stated explicitly so that these
results can be seen in their true limitations and ambiguities.
Although this clustering procedure was not without
problems,

its use could be justified on practical grounds:

in an ordinary factor analysis
groups analysis)

(as opposed to a multiple-

e ac h of the variables

(the 167 traits,

this study) would have a score for each of the factors

in
(the

1 High-loading departments are listed on pp. 69-71,
above.
High loadings indicate that a department has most
of its variance in common with the group, i.e., that the
department strongly resembles the group.

86

department g r o u p s ) ; the department groups w ould have to be
interpreted in terms of these 167 scores.

Since there are

no previous studies of these data, interpreting 167 traits
in order to describe the department groups would have been
inconvenient and impracticable.

By clustering the traits

b ef or e grouping the d e p a r t m e n t s , this study produced two

sets of loadings:

the traits'

clusters and the departments'

loadings on the traitloadings on the department

g r o u p s , and these loadings could be used in the interpre­
tation of the results of this study as well as for deter­
mining future studies of this topic.
These procedures seemed appropriate for an initial
attempt at a functional classification of university de p a r t ­
ments.

Further studies would probably benefit from using a

"standard" factor analysis

(such as a common factor analysis

using the principle axes technique,

in Q-analysis pattern,

followed by an oblique or orthogonal r o t a t i o n ) .

Although

the results of a factor analysis m ight be difficult to
interpret,

the task would hopefully be facilitated by

reference to the results of this study.

The clusters and

groups developed in this study represent an initial effort
to ascertain the dimensions of a terrain as yet inadequately
charted.

87

Problems from the Data
Like any empirical investigation,

this study could

be only as precise and meaningful as its data.

Unfortu­

nately the nature of the available data was not always
satisfactory.

The problem concerned the departmental traits,

both the manner in which some were reported and the lack of
reporting others.
This problem took three forms:

(1) In some

instances the traits did not reflect aspects of the situa­
tion that were apparent to common sense.
departments of Mathematics, Chemistry,

For example,

the

and Physics all had

high loadings on Group 3, although common sense suggested
that Mathematics does not have experimental laboratories or
high consumption of laboratory m a t e r i a l s , while Chemistry
and Physics do.

These aspects relevant to experimental

science departments were apparently not reflected in the
available data.

Specifically,

the data reported for labo­

ratory sections actually refer to a teaching method rather
than to instruction that takes place in experimental labo­
ratories.

(2) Some of the traits as reported actually could

have been divided into two or more traits;
instructional category "sub-college"

for example,

included both remedial

courses and short vocational-technical courses,

and the

category "independent study, variable credit" included
several forms of instruction.

the

88

In addition,

(3) some traits were interesting but

irrelevant or confusing.

For example,

the maleness traits

(student and faculty) were interesting since they related to
current social concerns; but their relevance to departmental
functions was not clear.

The interpretation of a depart­

ment's maleness was confounded because the maleness traits
clustered with the traits for faculty tenure and professor­
ial rank,
selves.

and therefore could not be considered in them­
This example indicates how the problems inherent

in the methodology and in the data at times reinforced each
other.
The trait-clusters that included maleness and new
or transfer majors were eliminated from descriptions of the
departmental groups,

although they were among the variables

on which the department groups had been formed.
A valid and meaningful classification of departments
according to functional traits depends upon adequate gather­
ing and reporting of data representing such traits.

Studies

such as the present one can indicate how the procedures for
handling these data could be improved.

Trait-Clusters
The procedures used in this study revealed sixteen
basic dimensions in the traits, or in other words, revealed
a pattern of interrelationships that included sixteen clus­
ters of traits.

These dimensions or clusters could also be

considered factors, as mentioned previously

(page 53).

89

The trait-clusters were relatively independent,
although several clusters correlated with other clusters.
In Figure 1, above, the intercorrelations formed three sub­
sets of trait-clusters:

1-2-3, 4-5-6,

6-7-8.

These corre­

lated sub-sets suggested that further factor analysis of the
trait-clusters would probably produce three or more secondorder factors reflecting broad categories such as curriculum,
expenditures,

and faculty time distribution.

Although some trait-clusters were correlated,

they

should be considered operationally independent for the
following reasons.

First,

the t r a it-clusters 1 alphas

(index

of internal consistency) were considerably higher than any
cluster's correlation with another cluster
Second,

(see Table 1).

the trait-clusters related independently to the

d e p a r t m e n t s ; for e x a m p l e , few departments had high scores
on these correlated trait-clusters as shown in Table 4.
An interesting result of the clustering process was
that certain trait-clusters that one might expect to show a
relationship actually did not correlate.

For example,

sponsored research funds and faculty time in research did
not correlate with the dimension of graduate curriculum;
also, faculty time in service functions did not correlate
with the dimension of graduate-professional curriculum.
These trait-clusters were specific to the concrete
situation at Michigan State University in 1969-1971.

The

90

Table 4.

Relation of Trait-clusters and Departments

Trait-Clusters

Number of
Departments with
High Scores on Both

1-2

5

2-3

7

1-3

0

4-5

9

5-6

2

4-6

0

6-7

1

7-8

1

6-8

7

6-7-8

1

traits in some cases reflect situations that were accidental
to a department's essential functions, e.g., classroom size,
laboratory stations, outside funds available,
matters external to a department.

and other

Such traits are acci­

dental in the sense that they are not usually essential to
a department's functions.

But in another sense, such e l e ­

ments are r e l e v a n t to a department's functions because
teaching and other activities necessarily occur in a
particular physical setting.

This specificity was appro­

priate to the study's objective of grouping departments in
terms of their actual functions in their specific context.

91

Department Groups
When the departments that loaded below .67 on the
department groups
eliminated,

(see list, pp.

69-73,

above) were

the department groups appeared slightly more

homogenenous and meaningful.
were more or less atypical

Some low-loading departments

(for various reasons)

department groups with which they correlated.

of the

For example,

the Department of Economics has functional similarities both
to social science departments and to some departments in the
College of Business,
department groups

to which it belongs.

(1, 2, 3, 4, 5,

6

In six of the

) the high-loading

departments were relatively coherent in subject matter, even
though some department groups included subject matters that
have traditionally been separated,

such as the natural sci­

ence subject matters that were combined in Group 3.

The

notable exception to this coherence was the Department of
Religion which was included in Group 2, Business and Engi­
neering.

In three of the department groups

(9, 10, 11) the

high-loading departments appeared to be dissimilar in both
subject matter and manner of operation.

It is worth noting

that in this sub-set of more heterogeneous groups, most of
the departments represented relatively new subject matters
(Advertising, C o m m u n i c a t i o n ) ,
nized colleges

or were in recently reorga­

(Human Nutrition and Foods),

or had recently

shifted position in the organizational structure
Justice,

Theatre).

(Criminal

92

Relation of Department Groups
and Colleges
One of the questions of interest in this study is,
h ow did the department groups produced by an analysis of
d e p a r t m e n t s 1 functional characteristics relate to the col­
lege structures at Michigan State University?

The answer

is that these particular department groups coincided with
the college structure to a greater degree than might have
been expected.
This finding can be illustrated in several ways:
First,

thirty-five departments

(44% of the population)

belonged to department groups in which the majority of
departments were from the same college as the department
in question,

e.g., Agricultural Economics belonged to

Group 1 in which the majority is also from the College of
Agriculture.

Second, in each group the percentage of

departments belonging to the same college was relatively
high for seven groups, as shown in Table 5.

Thus,

it was

evident that Groups 1 to 7 were relatively homogeneous,
Groups

8

to 11 relatively heterogeneous.
Third, for several colleges m ost of the component

departments correlated with only one departmental group, as
shown in Table

6

.

The figures in that table indicated that

the colleges of Agriculture, Arts and Letters,
Engineering,

Business,

and Veterinary Medicine were relatively

homogeneous in terms of their d e p a r t m e n t s ' functional

93

Table 5.

Departments in Each Group that Belonged to the Same College
(percents)

All Departments
in the Group

Department Group

High-Loading
Departments Only

(%)

(%)

1

80

100

2

42

44

3

50

60

4

78

100

5

60

50

6

100

100

7

100

100

8

—

—

9

—

—

10

—

—

11

—

—

characteristics

(those included in this s t u d y ) , while the

colleges of Communication A r t s , Natural S c i e n c e s , and Social
Sciences were relatively heterogeneous.
These findings supported the position expressed in
Chapters

I and II, above,

that organizing colleges on the

basis of departmental subject matters does not always com­
bine departments that function similarly.

These findings

suggested that in terms of their functions the departments
in certain colleges
Science)

(such as Natural Science and Social

at Michigan State University are similar to

departments in other c o l l e g e s .

Table 6.

Relation of Department Groups and Colleges, From College Perspective

Number of Departments in Each Group from Each College

College

Total Depts.
in College

1

2

Agriculture

13

12a

.

.

3

4

Group Number
5
6
7

8

9

10

.

.

■

1

•

1

Arts & Letters

9

1

7

Business

6

5

.

1

•

•

Comm. Arts

6

•

•

2

2

1

Engineering

6

4

•

1

1

•

Human Ecology

3

•

1

•

•

1

Human Medicine

3

•

•

3

•

•

•

Vet. Medicine

8

•

•

1

•

•

1
2

Natural Science
Social Science
Univ. College
Total

•

12

3

2

3

•

1

1

•

9

•

•

3

1

•

2

2

__4

•

•

•

•

•

4

•

•

•

12

6

9

5

4

7

6

5

79

15

6

aTwelve departments in Group 1 are also members of the College of Agriculture.

11

1

1

1

4

The newest college--Communication A r t s — appeared to
be one of the most heterogeneous colleges,

an observation

that raises the question of a college's effect on its
departments'

manner of operations.

Does membership in

a college over a period of time tend to make the component
departments more similar?

The example of the colleges of

Natural Sciences and Social Sciences indicated a negative
answer,

since their component departments are relatively

heterogeneous in the terms of this study.

Nevertheless,

perhaps a college's administrative practices do ultimately
affect its component departments.

A possible example of

such an effect could be the Department of Economics, which
loads high on Group 3 (Scientific Disciplines)

and corre­

lates especially high with Mathematics, but also correlates
high wit h departments in its own a o l l e g e
are in Group 2.

(Business) which

Perhaps, being in the College of Business

has had an effect on Economics.

One might compare the

Colleges of Agriculture, Arts and Letters,
Social Science,

Natural Science,

to investigate whether certain variables

describing the colleges'

administrative practices appear

have affected the functioning of the departments in those
colleges.

to

96

Departmental Typology
A second question of interest in this study is,
what kind of groups were produced by the analysis of
departmental functions?
The results of the clustering procedures that were
applied to the departments have already been presented in
two ways:

(1 ) as a listing of the members of the department

groups, as seen on pages 69-7 3, above;

and

(2) as a set of

summary descriptions of the department groups, pages 76-80,
above.

A third way of presenting these results would be as

a set of general department types which could represent the
department groups more concisely.

Such types would be

constructs in which the dimensions

(or factors)

in the

population of departments are given a concrete expression
in terms of their salient traits, as shown by the traitclusters on which the types had high or low scores.

Thus,

departmental types and groups should be considered two ways
of presenting the same information.

The eleven department

groups illustrated certain dimensions in the data,
categories within the department population.

i.e.,

A department

typology illustrates the same dimensions by summarizing the
characteristics in a reduced number of categories.
To simplify the interpretation of the department

i
I

groups,

it appeared desirable to amalgamate the groups

f

wherever possible.

This process was accomplished by

(1)

eliminating those departments whose group loadings were

&

97

below .67

(as shown on pp.

69-73,

above)

and

(2) by

combining those groups that had reasonably high correlations
(as shown in Figure 2).
The groups that were combined and their correlations
were as f o l l o w s :
Groups
Groups
Groups
Groups

2and 9
4and 7
5and 6
5and 10

(r=
(r=
(r=
(r=

.33)
.33)
.52)
.44)

When the pattern of relationships among the
departments and groups of departments had been clarified in
this way,

it became apparent that the departments formed

two broad classes according to:

(1 ) their instructional

emphasis, graduate and upper division instruction on the
one hand,

lower division instruction on the other;

their faculty size, distribution,

and work load.

and

(2 )

Each of

these classes included three department groups that showed
particular differences in curriculum emphases,

staff, spend­

ing, and funding patterns.
The list of department groups that resulted from
this revision and reordering was as f o l l o w s : 1

1 Group 10 disappeared, since three of its members
had group loadings below .67 and the remaining department
did not correlate well with any other group.
The Department
of Music was eliminated from Group F because its methods of
teaching (emphasis on instrumental practice) put it in a
class by itself.
The Department of Astronomy was eliminated
because it was atypical of B, although basically similar in
curriculum e m p h a s e s .

98

I.

GRADUATE ORIENTED DEPARTMENTS
Type A:

1.
3.
4.
5.
7.
8 .
9.
1 2 .
13.
51.
53.

Agricultural Sciences
Agricultural Economics
Animal Husbandry
Crop and Soil Science
Dairy
Food Science
Forestry
Horticulture
Poultry Science
Resource Development
Botany
Entomology

TY£e B:. Health Sciences
43.
45.
47.
50.
73.
74.
75.
76.
77.
78.
80.
Type C:
24.
32.
33.
39.
6 6 .
6 8 .
H .

Human Nutrition and Foods
Human Development
Psychiatry
Biophysics
Anatomy
Large Animal Surgery and Medicine
Medical Technology
Microbiology
Pathology
Pharmacology
Small Animal Surgery and Medicine
Service Technology
Business Law and Office Administration
Journalism
TV and Radio
Mechanical Engineering
Social Work
Urban Planning and Landscape Architecture

LOWER DIVISION ORIENTED DEPARTMENTS
Type D:
14.
16.
17.
18.
20.
22.
69.
70.
72.

Humanities and General Studies
Art
German and Russian
History
Linguistics
Philosophy
Romance Languages
American Thought and Language
Humanities
Social Science

99

Type E.
52.
55.
57.
65.
67.

Scientific Disciplines
Chemistry
Mathematics
Physics
Psychology
Sociology

Type F;
21.
23.
26.
27.
28.
35.
36.
37.
40.
58.
61.
64.

Applied Sciences and Technology
Religion
Accounting and Finance
Hotel, Restaurant, and Institutional Management
Management
Marketing and Transportation Administration
Chemical Engineering
Civil Engineering
Computer Science
Metallurgy, Mechanics, and Materials Science
Statistics
Criminal Justice
Political Science

One can describe the general types by relating the
relevant departmental traits to these revised g r o u p s , in the
same manner as had been used to describe the eleven depart­
m ent groups on pages 76-80, above

(see Table 7).

The Two General Classes of
Department Types
The chief distinguishing features of the two classes
of department types are
emphases and

(1 ) instructional or curricular

(2) faculty size and function.

In Class I,

instruction is mainly on the graduate and upper division
levels;

there is a r e l a t i v e l y 1 small faculty with much of

JIn this discussion the terms "relatively" and
"average" imply relationships among these 79 departments
at Michigan State University.

Table 7.

Departmental Typology

Departmental Types

A
Agricultural
Sci e n c e s

Ge n e r a l D e p a r t m e n t a l Tr a i t s

Curriculuma

Staff size

T e a ching
load

M o d e of
teaching

F a c u l t y time
distribution
E x p enditures

G r a d u a t e a n d upper d i v i s i o n
Lower d i v i s i o n
Gr a d u a t e - p r o f e s s i o n a l
Vocational-technica1 curriculum
F u l l - t i m e faculty
P a r t - t i m e f aculty
G r a d u a t e ass i s t a n t s
Faculty:
independ. study, v a r i a b l e credit
G r a d u a t e a s s i stants:
class
Faculty:
class
Grad, assistants:
independ. study,
variable credit
Nu m b e r o f courses, g r a d u a t e
Number o f courses, u n d e r g r a d u a t e
Small r e c i t a t i o n s e ctions
S t u d e n t s s eeking t e a c h i n g c e r t i f i c a t e s
R e c i t a t i o n me t h o d
L a b o r a t o r y me t h o d
La r g e lectures, und e r g r a d u a t e
I n d e p e n d e n t study, v a r i a b l e credit
Res e a r c h
Administration
Se r v i c e
Total
Ge n e r a l Fu n d

high
medium hiqh

larqe
**
**

I
C
S e r v i c e Or i e n t e d
Technology

m e d i u m hi g h
low
***
hi g h
below-average
below-average
larqe
**
**

H u m a n i t i e s and
G e n e r al Studies

C L A S S II
E
Scientific
Disciplines

F
A p p l i e d S c i ence
and Technology

low
high

low
high

medium high

D

m e d i u m high
m e d i u m high
low
m e d i u m hiqh
above-average

large
large
small

above-average
above-average

**

**

**

**

**

low

low

large*5
largec
small

above-averaqe
**
•*

**

m e d i u m hi g h
high
high
high

m e d i u m high

...
m e d i u m high

m e d i u m high

m e d i u m hi q h
m e d i u m high
m e d i u m hi g h

Pe r c e n t for labor
Pe r c e n t for service, supplies, e q u ipment
Sp o n s o r e d r e s e a r c h
Ex p e r i m e n t S t a t i o n a n d / o r E x t e n s i o n
Ed u c a t i o n a l D e v e l o p m e n t P r o g r a m

Support
funds

m e d i u m high
*

CLASS
B
Health
Sciences

m e d i u m high
m e d i u m high
high

m e d i u m hi g h
hi g h
***

medium
medium
medium
medium

high
high
high
high
m e d i u m high
low

high
high
hiqh
high

hiqh

low
m e d i u m hi g h
hiqh

high
high
high
high
high
high
hiqh
high

...
m e d i u m high
medium high
m e d i u m high

t
high
high
...

ft
m e d i u m high

high

low

hi g h
high
m e d i u m high

Rep r e s e n t s number of courses, c o u r s e enrollments, tea c h i n g
credits, student c r e d i t hours.
Do e s n o t ind i c a t e d e p a r t m e n t a l
prestige, q u a lity or importance.

hi g h

low

*No en t r y

(...)

s i g n i f i e s that trait is p r e s e n t to an

a v e r a g e degree.
• • S i g n i f i e s that

trait

is present.

b
Small
c

in d e p a r t m e n t s

E s p e c i a l l y large

in U n i v e r s i t y College.

in d e p a r t m e n t s

m

U n i v e r s i t y College.

••• H i g h for d e p a r t m e n t s in C o l l e g e of V e t e r i n a r y M e d i c i n e only.
+ H i g h in P s y c h o l o g y only.
t t H i g h in C r i m i n a l J u s t i c e only.

1Q1

its work load in independent study and relatively many
graduate assistants who teach classes.

In Class II,

instruction is mainly on the lower division level; there is
a relatively large faculty

(both full-time and part-time)

with a medium-to-high work load in classes, and a relatively
few graduate assistants who handle independent study and
variable credit classes.
No clear pattern of differentiation between the two
main classes is apparent with respect to the other general
departmental traits— mode of teaching,
bution, expenditures,

faculty time distri­

and sources of funds— w hich serve to

differentiate the departmental types.

The Six Basic Types of Departments
Type A;

Agricultural S c i e n c e .

This type is char­

acterized by instruction at the graduate and upper division
levels, with an emphasis on graduate-professional curriculum.
It offers many courses at all levels.

Its faculty size is

average, and it has many graduate assistants.

The faculty

tends to handle the independent study and variable credit
courses, and the graduate assistants teach the classes.

The

faculty spends a relatively high percent of its time both in
research and administration.

It receives large support from

the Experiment Station and Extension Service.
tures for labor are relatively large.

L

Its expendi­

102

Type B;

Health S c i e n c e .

In this type, instruction

is chiefly on the graduate and upper division level, with
relatively little instruction on the lower division level.
It offers many short vocational-technical courses
as "sub-college").

(classed

The faculty is relatively small, and

the number of graduate assistants large.

The faculty tends

to work in independent study, and the graduate assistants
teach classes.

Type B includes two sub-types:

medicine type and a human medicine type.

a veterinary

At Michigan State

University,

the human medicine type is new and in a state of

developing,

a situation that apparently results in a large

amount of faculty time spent in administration, as well as
resulting in relatively low total spending.

The veterinary

medicine sub-type offers a large graduate-professional
curriculum.

Type B has high sponsored research support.

Type C:

Service-Oriented T e c h n o l o g y .

In Type C,

instruction is offered about equally on both the upper
division-graduate and the lower division levels.

The size

of both its full-time faculty and its graduate assistants
staff is above average.

The faculty's work load is gener­

ally in independent study, and the graduate assistants'
is chiefly in classes.

The faculty spends a high amount of

its time in service fun c t i o n s , and more than the average
time in research.

load

103

Type D;

Humanities and General S t u d i e s .

In this

type, instruction is offered chiefly at the lower division
level, and is relatively less prominent at the upper d i v i ­
sion and graduate levels.

Its faculty is large— both full­

time and part-time— and it has few graduate assistants.
Faculty work load is medium-heavy in classes; graduate
assistants handle independent study.

Its faculty spends a

relatively high percent of its time in research, but less
than average time in administration.

Type D is represented

by two slightly different sub-types:

the humanities type

and the general studies type.

The latter type differs by

having a larger part-time faculty and a smaller full-time
faculty.
Type E:

Scientific Di s c i p l i n e s .

Instruction for

Type E is for the most part similar to that offered by
Type D, although Type E offers more undergraduate courses.
Faculty and staff size and function is also similar to that
of Type D.
research.

Faculty in Type E spend a high amount of time in
Expenditures are high for Type E.

Type F:

Applied Science and T e c h n o l o g y .

Type F

offers slightly more instruction on the lower division level
than on the upper division and graduate level.
is larger than average

Its faculty

(but relatively smaller than that of

Type D or Type E) and it has relatively few graduate assis­
tants

(but more than Type D or Type E ) .

As in the other

104

types in Class II, in Type F the faculty has a medium-toheavy work load in class, and the graduate assistants work
chiefly in independent study and variable credit courses.

Subject Matters of the Departmental
Types
It might appear that the subject matter of each
department type is a coherent and reasonable selection from
the full spectrum of university studies.
not actually that simple, however.

The situation is

For each type, one can

invent reasons why its subject matters fit together, and
yet it is clear that one could find other equally good
reasons for several different distributions of the segments
(i.e., departmental subject matters) among the six types.
Agricultural Sciences,

for example,

appears to be a coherent

group of studies; but, three of its segments deal with
animal management and could conceivably be grouped with
animal medicine;

four other segments deal with plants and

could form their own subject matter group; Agricultural
Economics and Resource Development could be combined with
Economics and even Management; while Food Science could join
with Human Nutrition and Foods.

A similar analysis could be

applied to the subject matters within each of the other five
types.

The two technology types

(C and F) quite clearly

indicate diverse segments of subject matter, but even in
those that appear more homogeneous— Humanities or Scientific

105

Disciplines— the segments could be broken apart and
reassigned according to several relational schemes.
The point to be made is that these six categories
of subject matters were formed not be c au se of any supposed
necessary relationships among them, but in spi te of their
dissimilarities.

The six combinations of segments of

studies coincide to some extent with the actual colleges in
this university, but the departmental types coincide with
the usual subject matter divisions no more

(or no less)

than

the colleges do.
The grouping of subject matters that resulted from
the procedures of this study was based on similarities in
the d e p a r t m e n t s ' f u nctions.

The fact that some of the sub­

ject matters group together probably occurred because some
of the departments have been grouped together in a college
and have over a period of time developed in similar manner.
The similarities among departments mos t likely reflect
a particular historical situation,

an assignment to a

particular college for various reasons of administrative
convenience.
Thus the grouping of subject matters into six
departmental types reflects in some instances an actual,
particular organization of departments into colleges,
to the extent that this situation obtains

and

(and to the extent

that the college structure is based on reasons other than
departments'

real similarities), the grouping of subject

106

matters is based on accidental rather than essential
characteristics.
The ultimately inconsistent way in w hich subject
matters are combined is also demonstrated by the variety
of kinds of subject matters in the six departmental types.
Type A

(Agricultural Science)

and Type B

(Health Science)

represent areas of application which involve several tech­
nologies and sciences.
Type F

Type C (Service Technology)

(Applied Science and Technology)

nificantly to various off-campus

and

both relate sig­

(non-university) clienteles

and are concerned with applications of various technologies
and sciences to areas that are defined in terms of the current
culture rather than in terms of natural processes.
(Humanities and General Studies)
Disciplines)

and Type E

Type D

(Scientific

are both concerned chiefly with the theoretical

aspects of their areas of study, but differ in methodology,
Type D usually employing analytical methods and Type E
employing experimental m e t h o d s .
The significance of these observations is this:
even groups of subject matters that appear reasonably
coherent and consistent,

such as those in the six depart­

mental types, may actually be organized according to several
different principles
of inquiry)

(area of application,

clientele, method

and separated according to other principles

(level of instruction,

faculty structure)

that are irrele­

vant to the essential aspects of subject matters.

107

Useful and meaningful groupings of departments
should be based on careful analysis of their actual sim­
ilarities.

Since these similarities would include those

in subject matter as well as in function

(and other aspects

as discussed in Chapter 1, above), no single organizational
structure is likely to suffice.

Summary
1.

The method used in this study was useful for

achieving the study's objectives, but somewhat inconvenient
for interpreting the results.

The most acute problem was

that many traits were "lost" within the trait clusters.
2.

The available data were not completely adequate

to express departmental functions, and therefore limited the
validity of the results.

Before a truly valid functional

classification of departments can be achieved,
ments'

the depart­

functional traits must be more adequately identified,

defined,

and recorded.
3.

The department groups coincided with the col­

leges at Michigan State University to a greater degree than
might have been expected.

Seven of the groups had a major­

ity of members from a single one of the colleges.

This

relation suggests that the effect of a college's adminis­
trative practices upon its component departments should be
investigated.

108

4.
above

After amalgamating the groups that correlated

.33 and eliminating low-loading departments from their

groups, one can discern a basic typology of departments.
The departments formed two main classes, and each class
included three types of department.
tinguished according to
courses, enrollments,
staff

The classes were d i s ­

(1 ) instructional mission

credit hours)

and

(size, distribution, work load).

upper division and graduate instruction;
lower division instruction.
were:

Class I emphasized
Class II emphasized

(B) Health Sciences,

(D) Humanities and General Studies,

(E) Scientific Disciplines,
Technology.

(2 ) faculty and

The six types of department

(A) Agricultural Sciences,

(C) Service Technology,

(level of

and

(F) Applied Sciences and

The departmental types differed on instruc­

tional emphasis,

faculty and staff

(size and function),

pattern of expenditures, and teaching practices.

The

fundamental distinguishing features of this typology are
departmental m i s s i o n and r es o u r c e d e p l o y m e n t .
5.

The subject matters of the departmental types

appeared to be reasonably coherent segments of university
studies, but that was chiefly because the types coincided to
some degree with the university's colleges and the colleges
also appeared to be coherent in subject matter.
the types

(and the colleges)

were not necessarily similar,

In fact,

included subject matters that
and the types represented at

least three different ways of defining subject matter.

CHAPTER VI
CONCLUSIONS

Evaluation of the Procedures
and Data
The procedures used in this study were ultimately
unsatisfactory.
(pp. 81-87)

As has been pointed out previously

the clustering of departmental traits,

although

its results were interesting in themselves, did not provide
a satisfactory base for grouping the departments.
However,

the procedures were fruitful in the sense

of demonstrating that departments can be grouped according
to their functions, and that the department groups produced
can be meaningful in terms of a common sense understanding
of the university.

These procedures also produced clusters

of traits that indicated the dimensions of this particular
set of departmental traits.
Ascertaining the loadings of individual traits on
their trait-clusters might prove a helpful contribution to
further studies of departments and their inter-relationships.
These loadings indicated which traits were salient within
the clusters and thus could be used to reduce the number of
traits used in further applications of procedures similar

110

to those in the present study.

Thus, although clustering

the traits prior to grouping the departments was not in the
final analysis an appropriate or satisfactory procedure,

the

clustering process was nevertheless useful and fruitful in
terms of w hat it has revealed about the interrelations of
the traits themselves.
The data available for this study were also
unsatisfactory in some instances, as has been pointed out
previously

(page 87).

This situation resulted partly from

the manner of gathering and reporting certain departmental
traits.

The traits that were available as data were

originally defined for use by various university units
(Office of Institutional Research,

Provost)

and not in terms

of the objectives of the present study and therefore a cer­
tain dysfunctionality in the data was to have been expected.
As part of the data,

the department population was

also somewhat unsatisfactory since the five departments of
the College of Education could not easily be included.

Attainment of Objectives
In the sense that the present study accomplished a
clustering of departmental traits and a grouping of depart­
ments,

its objectives were attained.

However, merely

attaining the objectives might not be sufficient for a real
evaluation.

In Chapter III, it was suggested that the suc­

cess of the present study should be judged according to the

Ill

precision, meaningfulness,
groups produced,

and utility of the department

and also by the efficiency and heuristic

value of the p r o c e d u r e s .
In terms of internal consistency and external dis­
creteness,
precise,

the eleven department groups were reasonably

and therefore the procedures were at least par­

tially successful.

The groups'

a l p h a s 1 were all high, and

the correlations among groups mostly low.
ment "types"

The six depart­

(formed by combining several groups with

relatively high correlations and by eliminating departments
w ith low loadings on their groups)

can also be assumed to be

internally consistent and discrete.
In terms of meaningfulness,
were less successful.

the grouping operations

One problem was that the department

g r o u p s ' scores on one of the trait-clusters were inexplica­
bly at odds with the actual facts of the situation.

Also,

as mentioned previously, the department groups could not be
fully characterized because the traits embedded in some of
the trait-clusters were excessively diverse.

Only when the

focus was on the subset of traits related to the depart­
ments'

instructional missions and staff deployment could the

six types be related with a meaningful pattern.

Thus,

the

ultimate meaning of the groups was the result of a certain

1Alpha is a measure of internal consistency.
footnote 1, p. 46 and Table 2, p. 67.

See

112

amount of manipulation and interpretation.

However, both

the eleven groups and the six types appeared to reflect
observable situations in the university to a considerable
degree, and had meaning to that extent.
The utility of the groupings— as opposed to the
procedures— is questionable.

This set of groups would not

likely be helpful for evaluating or budgeting departments,
because the groups are not clearly enough based on coherent
sets of traits.

To be useful, department groups should be

based on the dimensions within subsets of traits that are
relevant to more specifically defined departmental functions,
such as budgeting,

staffing,

instruction.

Only in terms of the paths it might suggest for
other investigators, can the present study be considered
to have been worth its investment.

Hopefully,

its potential

or heuristic value will outweigh its actual or practical
value.

On one path of inquiry,

future investigators might

be encouraged to use similar methods of analysis on data
gathered at other institutions to ascertain whether the
faults of the procedures in the present study might be
avoided.

On another path,

investigators might seek to use

other forms of factor analysis to group departments on the
basis of various subsets of data.

Perhaps the most

heuristic effect of the present study will have been its
attempt to focus some research interest on the problems
of the organizational structures of universities.

113

Significance
This study has illustrated how the functional
aspects of university departments— their instructional
m i s s i o n — can provide a basis for classifying departments.
An organizational system based on functional traits could
replace or be added to the systems used at present in most
universities,

in which departments— arbitrarily defined

according to principles that are "historical" and therefore
non-rational— are conjoined into colleges or divisions that
are also non-rational.

These arbitrary defining principles,

in their most basic form, are usually assumptions about
academic subject matters,

stipulating points of segmentation

in the universal body of knowledge.

This way of defining

departments and organizations of departments may be no more
or less arbitrary than other assumptions about the structure
of knowledge, but the structures based on such assumptions
are nevertheless inadequate for achieving the university's
purposes at the present time.
There are two major reasons why most one-dimensional,
subject-matter-based organizational structures in contempo­
rary universities are inadequate.

First, a one-dimensional

or one-principle structure cannot adequately reflect the
complex reality of a university.

This deficiency has of

course been present in such structures from the start, but
has become increasingly apparent— and dysfunctional—

as the

university and all of its components have grown more complex.

114

When all the essential aspects of a department in a modern
university are considered,

the possibilities for variation

are so great that no system based on a single principle
could adequately define a superstructure for the departments.
This problem has at least two kinds of structural
solutions:

one, to develop new superstructures in which the

existing departments are conjoined in several different
patterns according to different objectives;

the other,

to

split up the department as it is generally known today into
its various aspects and develop new kinds of units to serve
the particular aspects.
The present study relates to the first solution by
illustrating a superstructure in which existing departments
are conjoined on the basis of their functions,
instructional mission.

chiefly their

Other superstructures could conjoin

departments according to other departmental a s p e c t s ; for
example, on the basis of subject matters, departments could
be organized into the usual three or four divisions, or into
humane studies and environmental studies, and so on; on the
basis of resources, departments could be organized into
groups with different patterns of spending and funding.
As for the second solution— new units— the present
study illustrates how instructional missions could be used
to define six units that could then be further divided into
sub-units according to specific missions

(variations p r o ­

duced by the interaction of a general mission with other

115

departmental aspects (subject matter,
and resources).

students,

faculty,

A t the same time, other units could be

established to serve objectives defined in terms of the
other aspects

(memberships, resources,

and so on);

for

example, socialization of students as professionals could
be handled by units specialized according to major voca­
tional areas

(business administration,

research,

teaching,

engineering)

separate from the instructional units although

staffed by some of the same faculty.
A third kind of solution to this problem

(i.e., the

inadequacy of present-day departmental structures to reflect
the complex reality of the university) would be theoretical
rather than structural:

to study the department's charac­

teristics and relationships in order to define its dimen­
sions better.

As a step toward such a solution the present

study has attempted to investigate the dimensions within a
set of departmental traits that chiefly represent functional
aspects.

However,

the traits included in the present study

represent only a limited part of a department's functions.
Other functions such as those relating to student socializa­
tion and faculty career management must also be analyzed for
a clearer understanding of the complexity of the department.
The second reason why the usual structures for
organizing departments are inadequate is that they hamper
evaluation of departments and therefore impede improvement
of departments'

performance.

At present, neither university

116

administrators nor education officials and critics outside
the university have any valid ways of evaluating departments.
Almost nothing can be stated with assurance about the key
characteristics or dimensions of departments in American
universities.
planners,

In the terms often used by contemporary

the input and o ut p u t of departmental operations

cannot be clearly stated.

(Should the faculty's increased

experience in teaching or research be considered an output
of a department's operations?)

Not only are departments'

input and output ill-defined or even undefinable but the
relations between them are virtually impossible to assess
and evaluate.

Such relating concepts as "efficiency,"

"value added,"

"cost-benefit" have not yet been given

meanings that are applicable or even relevant to educational
operations.

It is impossible to state what "efficiency"

might mean for departmental operations

(for example, effi­

ciency in the production of student credit hours, or effi­
ciency in the ratio of students to faculty)
"acceptable"

level of efficiency,

or to state an

in the absence of empir­

ically valid generalizations about university departments.
Valid generalizations about entities as complex as
departments and their functions cannot be developed without
more extensive observations and hypotheses than exist at
present,

and without greater knowledge of the true simi­

larities among departments.
scientific)

Since adequate

(that is,

criteria for evaluating departments are not

117

only lacking at present but also appear to be unlikely
developments in the near future, comparison is and will
remain for some time a basic tool for evaluation.

But at

this point the entire evaluation process founders,

since in

our present state of knowledge we have no sound way of tell­
ing the apples from the oranges

(not to mention the lemons)

among the departments.
It is in the light of these considerations that the
one-dimensional,

subject-matter-based structures in which

most universities organize their departments appear inade­
quate for the evaluation and subsequent improvement of
departments.

Such structures have most likely inhibited

the normal human tendency towards comparisons by juxtapos­
ing departments in subject matter groupings with easily
perceivable functional dissimilarities, and by hiding
similarities that actually exist, and— perhaps the most
harmful effect--by continually emphasizing in title and
practice that subject matter is the most important aspect
of departments and their superstructures.
The overall effect is an attitude that disdains both
comparisons of departments and a broad conception of depart­
mental functions, an attitude that has been amply reflected
in the conservatism and anti-rationalism of university
people— faculty especially but also administrators— when
concerned with their own operations.

118

The present study has been conceived as part of a
propaedeutic to the evaluation and improvement of depart­
ments.

It has been intended as a study of the sort of

procedures that must be applied and the sort of data that
must be gathered before an evaluation of departments can
be achieved.

BIBLIOGRAPHY

T
BIBLIOGRAPHY

Ashby, Eric.
"On Universities and the Scientific Revolution,"
1958.
E d uc at io n, E c o n o m y , a n d Society.
Edited by
A. H. Halsey, Jean Floud, and C. A. Anderson.
Glencoe, 111.:
The Free Press, 1961, pp. 466-476.
Baldridge, J. Victor, ed.
A c a d e m i a Go ve rnance.
Berkeley,
Calif.:
McCutchan Publishing Corporation, 1971.
Bereiter, Carl, and Mervin B. Freedman.
"Fields of Study
and the People in Them."
The A m e r i c a n College.
Edited by Nevitt Sanford.
New York:
John Wiley
& Sons, Inc., 1967, pp. 563-596.
Bess, James L.
"Ranking Academic Departments:
Some
Cautions and Recommendations."
J o u r n a l o f H ig h e r
E d u c a ti o n, 42, 9 (December, 1971), 721-727.
Burt, C. L.
"The Appropriate Uses of Factor Analysis and
Analysis of Variance."
Handbook of Multivariate
E x p e r i m e n t a l P s y c h o l o g y . Edited by R. B. Cattell.
Chicago, 111.:
Rand McNally & Company, 1966.
Cattell, R. B.
M an ua l

Fa c to r A n a l y s i s : An I n t r o d u c t i o n an d
f or the P s y c h o l o g i s t a nd S o c i a l S c i e n t i s t .

New York:

Harper and Row,

19 56.

_, ed.
Handbook of Multivariate Experimental P sy ch o l o g y .
Chicago, 111.:
Rand McNally & Company, 1966.
Clark, Burton R.

"Academic Society, Dynamics of."
of Education.
1, 34-39.

Encyclopedia

Clark, Terry N.
"Institutionalization of Innovations in
Higher Education:
Four Models."
Academic Gover­
nance.
Edited by J. Victor Baldridge.
Berkeley,
Calif.:
McCutchan Publishing Corporation, 1971,
pp. 75-96.
Corson, John J.
G o v e r n a n c e o f C o l le g es an d U n i v e r s i t i e s .
New York:
McGraw-Hill & Company, 19 60.

119
e

120

Currie,

I. D., H. C. Finney, T. Hirschi, and H. C. Selvin.
"Images of the Professors and Interest in the
Academic Profession," 1966.
Sociology of Education
Edited by Ronald M. Pavalko.
Itasca, 111.:
F. E.
Peacock Publishers, Inc., 1968, pp. 532-554.

"The Embattled University."
Proceedings of the American
Academy of Arts and Sciences.
D a e d a l u s , 99, 1
(Winter, 1970).
Dressel, P. L . , F. C. Johnson, and P. M. Marcus.
The
C on f id e n c e C r i s i s . San Francisco, Calif.:
Jossey-Bass, Inc., Publishers, 1970.
Fruchter, B.
Introduction
VanNostrand, 19 54.

New York:

to F a c t o r A n a l y s i s .

Gaff, J. C., and R. C. Wilson.
"Faculty Cultures and
Interdisciplinary Studies."
J ou r n a l o f H i g h e r
E d u c a ti o n, 43, 3 (March, 1971), 186-201.
Gottlieb, David.
"Processes of Socialization in American
Graduate Schools," 1961.
S o c i o l o g y o f Education.
Edited by Ronald M. Pavalko.
Itasca, 111.:
F. E.
Peacock Publishers, Inc., 1968, pp. 258-270.
Haas, Eugene, and Linda Collen.
"Administrative Practices
in University Departments," 1963.
Academic
G o v e r n a n c e . Edited by J. Victor Baldridge.
Berkeley, Calif.:
McCutchan Publishing Corporation
1971, pp. 193-207.
Halsey, A. H.
"The Changing Functions of Universities,"
1960.
E d u c a t i o n , E c o n o m y , a n d Society.
Edited by
A. H. Halsey, Jean Floud, and C. A. Anderson.
Glencoe, 111.:
The Free Press, 1961, pp. 456-465.
Halsey, A. H . , Jean Floud, and C. A. Anderson.
E c o n o m y } and Society.
Glencoe, 111.:
Press, 1961.

Eduaation3

The Free

Harman, H. H.
M od e r n F ac t or A n a l ys i s.
Rev. ed.
University of Chicago Press, 1967.

Chicago:

Hunter, J. E.
"Arrange:
A Method of Reordering the
Correlation Matrix."
Paper presented at the
Midwestern Society of Multivariate Experimental
Psychology, May, 1970.

121

Hunter, J. E., and W. J. Browne, Jr.
"A Cluster Analysis
of The Marlowe Crowne Social Desirability Scale."
Paper presented at Midwestern Society of M u l t i ­
variate Psychology, May, 1971.
Hunter, J. E., and S. H. Cohen.
"Package:
A System of
Computer Routines for the Analysis of Correlational
Data."
Educational and Psychological Measurement,
29, 3 (Autumn, 1969), 697-700.
Ikenberry,

Stanley 0.

Journal

"The Organizational Dilemma."
43, 1 (January, 1972),

of Higher Education,

23-34.
Jencks, Christopher,
Revolution.

and David Riesman.
The A c a d e m i c
New York:
Doubleday and Company,

I n c ., 1969.
Joughin, Louis.

"Faculty, College and University."
o f E d u c a t i o n , 3, 494-503.

Encyclopedia

Kerr, Clark.
"Governance and Functions."
1 (Winter, 1970), 108-121.
Kiger, Joseph C.
"Disciplines."
3, 99-105.

D a e d a l u s , 99,

Encyclopedia of Education,

Kim, Richard C. C.
"The Academic Division:
to Excellence?"
Liberal Education,
1966), 428-433.

Stumbling Block
52, 4 (December,

McNamara, James F.
"Mathematical Programming Models in
Educational Planning."
Review of Educational
R e s e a r c h , 41, 5 (December, 1971), 419-446.
Moran, W. E.
"Measurement of Decentralization in University
Organizations."
American Educational Research
J ou r n a l , 8, 2 (March, 1971), 203-219.
Murray, Robert K.

"On Departmental Development:
A Theory."
o f G ene ra l E d u c a t i o n , 16, 3 (October,
227-237.

The J o u r n a l

1964),

Nunnally, J u m C.
P s y c h o m e t r i c Theory.
Hill, 1967.

New York:

McGraw-

Parsons, Talcott.
"The Strange Case of Academic O r g a n i z a ­
tion."
J o u r n a l o f H i g h e r E d u c a t i o n . 42, 6 (June,
1971), 486-495.
Pavalko, Ronald M . , ed.
S o c i o l o g y o f Ed uc at i on .
Itasca,
111.:
F. E. Peacock Publishers, Inc., 1968.

122

Rosen, Bernard C., and Alan P. Bates.
"The Structure of
Socialization in Graduate School," 1967.
Sociology
o f E d u c a t i o n . Edited by Ronald M. Pavalko.
Itasca,
111.:
F. E. Peacock Publishers, Inc., 1968,
pp. 244-258.
Rummel, R. J.
A p p l i e d F a c t o r Analysis.
Evanston,
Northwestern University Press, 1970.
Sanford, Nevitt, ed.
The A m e r i c a n College.
John Wiley & Sons, Inc., 1967.

111.:

New York:

Schuessler, K., and H. Driver.
"A Factor Analysis of
Sixteen Primitive Societies."
American S oc io ­
logical Review, 21 (1956), 493-499.
Selvin, Hanan C., and Warren O. Hagstrom.
"Determinants
of Support for Civil Liberties," 1960.
Sociology
of E d u c a t i o n . Edited by Ronald M. Pavalko.
Itasca,
Illinois:
F. E. Peacock Publishers, Inc., 1968,
pp. 155-178.
Spurr,

Stephen H.
"Organization of the Liberal Arts in the
Large University."
J ou rn al o f H i g h e r E du ca ti o n, 39,
1 (January, 1968), 11-22.

Vreeland, Rebecca S., and Charles E. Bidwell.
"Classifying
University Departments:
An Approach to the Analysis
of Their Effects Upon Undergraduates' Values and
Attitudes," 1966.
S o c i o l o g y o f Ed uc at i on .
Edited
by Ronald M. Pavalko.
Itasca, 111.:
F. E. Peacock
Publishers, Inc., 1968, pp. 179-197.

APPENDIX A

DEPARTMENTS THAT MADE UP THE POPULATION
OF INTEREST

APPENDIX A

DEPARTMENTS THAT MADE UP THE POPULATION
OF INTEREST

College of Agriculture and Natural Resources
.
.
3.

1

2

4.
5.
6

.

7.
8 .
9.
1 0 .
1 1 .
1 2 .
13.

Agricultural Economics
Agricultural Engineering
Animal Husbandry
(Biochemistry in College of Natural Science)
Crop and Soil Science
Dairy
Fisheries & Wildlife
Food Science
Forestry
Horticulture
Packaging
Park & Recreation Resources
Poultry Science
Resource Development

College of Arts and Letters
14.
15.
16.
17.
18.
19.
2 0 .
2 1 .
2 2 .

Art
English
German and Russian
History
Linguistics— Oriental & African Languages
Music
Philosophy
Religion
Romance Languages

123

124

College of Business
23.
24.
25.
26.
27.
28.

Accounting & Financial Administration
Business Law and Office Administration
Economics
Hotel, Restaurant, and Institutional Management
School of
Management
Marketing and Transportation Administration

College of Communication Arts
29.
30.
31.
32.
33.
34.

Advertising
Audiology & Speech Science
Communication
Journalism, School of
T.V. and Radio
Theatre

College of Engineering
35.
36.
37.
38.
39.
40.

Chemical Engineering
Civil and Sanitary Engineering
Computer Science
Electrical Engineering & Systems Science
Mechanical Engineering
Metalurgy, Mechanics, and Materials Science

College of Human Ecology
[41.
42.
43.
44.

Family E c o l o g y ] 1
Family and Child Sciences
Human Nutrition and Foods
Human Environment and Design

Eliminated.

125

College of Human Medicine

45.
46.

47.

(Anatomy in Veterinary Medicine)
(Anthropology in Social Science)
(Biochemistry in Natural Science)
(Biophysics in Natural Science)
Human Development
(Medical Technology, School of. in Veterinary Medicine)
Medicine
(Microbiology and Public Health in Veterinary Medicine)
(Pathology in Veterinary Medicine)
(Pharmacology in Veterinary Medicine)
(Physiology in Natural Science)
Psychiatry
(Psychology in Social Science)
(Sociology in Social Science)
(Zoology in Natural Science)

College of Natural Science
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.

Astronomy
Biochemistry
Biophysics
Botany and Plant Physiology
Chemistry
Entomology
Geology
Mathematics
(Microbiology and Public Health in Veterinary Medicine)
Nursing, School of
Physics
Statistics
Zoology

College of Social Science
60.
61.
62.
63.
64.
65.
6 6 .
67.
6 8 .

Anthropology
Criminal Justice, School of
Geography
Labor and Industrial Relations
Political Science
Psychology
Social Work, School of
Sociology
Urban Planning and Landscape

126

University College
69.
70.
71.
72.

American Thought and Language
Humanities
Natural Science
Social Science

College of Veterinary Medicine
73.
74.
75.
76.
77.
78.
79.
80.

Anatomy
Large Animal Surgery and Medicine
Medical Technology
Microbiology and Public Health
Pathology
Pharmacology
Physiology
Small Animal Surgery and Medicine

APPENDIX B

DEPARTMENTAL TRAITS THAT MADE UP THE DATA

APPENDIX B

DEPARTMENTAL TRAITS THAT MADE UP THE DATA

Description

No.

Source

Number of courses per course l e v e l :
number of separate courses taught,
fail, 1970.
1.
2.
3.
4.
5.
6 .

Number
Number
Number
Number
Number
Number

of
of
of
of
of
of

courses:
courses:
courses:
courses:
courses:
courses:

The

Sub-college, 001-009
Lower division, 100-299
Upper division, 300Graduate-professional
Masters
Doctors

OIR Report:
Volume of
Instruction
Analysis
It

Course e n r o l l m e n t s : The number of students enrolled in the various c o u r s e s ,
as of the tenth day of classes, fall,
1970
7.
8
9.

.

10 .
11
12 .

.

Course
Course
Course
Course
Course
Course

enrollments:
enrollments:
enrollments:
enrollments:
enrollments:
enrollments:

Sub-college
Lower division
Upper division
Graduate-professional
Masters
Doctors

Class hours, per course l e v e l : The number
of hours per week of instruction in organized classes , fall, 1970
13.
14.
15.
16.
17.
18.

Class
Class
Class
Class
Class
Class

hours:
hours:
hours:
hours:
hours:
hours:

Sub-college
Lower division
Upper division
Graduate-professional
Masters
Doctors

127

II
It
II
II
It
II

128

Description

No.

Source

Teaching credits in classes per course
l e v e l ; The number of credits of
teaching in organized, fixed-credit
classes, fall, 1970.
19.
2 0 .
2 1

.

2 2

.

23.
24.

Teaching credits
Teaching credits
division
Teaching credits
division
Teaching credits
professional
Teaching credits
Teaching credits

0 1 R Report:
Volume of
Instruction
Analysis
tv

in c l a s s e s :
in c l a s s e s :

Sub-college
Lower

in c l a s s e s :

Upper

in c l a s s e s :

Graduate-

in c l a s s e s :
in c l a s s e s :

Masters
Doctors

II
II
II

Teaching credits in independent study

II

II
II

l e v e l : The number of credits of teaching
in independent study and variable credit
courses.
25.
26.
27.
28.
29.

Teaching
credit
Teaching
credit
Teaching
credit
Teaching
credit
Teaching
credit

credits
classes:
credits
classes:
credits
classes:
credits
classes:
credits
classes:

in ind. study-var.
Sub-college
in ind. study-var.
Lower division
in ind. study-var.
Upper division
in ind. study-var.
Masters
in ind. study-var.
Doctors

Student credit hours, per course l e v e l :
The total number of credits for which
students are registered, as of the
tenth day of class, fall, 1 9 7 0 . 1
30.
31.
32.
33.
34.
35.

Student
Student
Student
Student
Student
Student

credit hours:
credit hours:
credit hours:
credit hours:
credit hours:
credit hours:

Sub-college
Lower division
Upper division
Graduate-professional
Masters
Doctors

‘For one course, the number of students enrolled
multiplied by the credit value of the course.

129

No.

Description

Source

Sections by type and s i z e :

36.
37.
38.
39.
40.
41.
(42.
43.
(44.
45.
46.
47.
48.
49.
(50.
51.
52.
53.
54.
55.
56.
(57.
(58.
(59.
(60.
(61.
62.
63.

fall,

1970

OIR Report:
Section Size
Analysis

Undergraduate lecture section:
201+
Undergraduate lecture section:
101-200
Undergraduate lecture section:
51 - 1 0 0
Undergraduate lecture section:
36 -50
Undergraduate lecture section:
2 1 -35
Undergraduate lecture section:
11 -20
Undergraduate lecture section:
1 -1 0 )1
Undergraduate recitation s e c t i o n : 2 0 1 +
Undergraduate recitation section:
1 0 1 -2 0 0 )
Undergraduate recitation section:
51-100
Undergraduate recitation section:
36-50
Undergraduate recitation section:
21-35
Undergraduate recitation s e c t i o n : 1 1 - 2 0
Undergraduate recitation section:
1-10
Undergraduate laboratory section:
2 0 1 +)
Undergraduate laboratory section:
101-200
Undergraduate laboratory section:
51-100
Undergraduate laboratory section:
36-50
Undergraduate laboratory section:
21-35
Undergraduate laboratory s e c t i o n : 1 1 - 2 0
Undergraduate laboratory section:
1-10
Graduate section:
201+)
Graduate section:
101-200)
Graduate section:
51-100)
Graduate section: 36-50)
Graduate section: 21-35)
Graduate section: 11-20
Graduate section: 1-10
Student credit hours, total per section
t y p e : The number of student credit
hours taught in section of the given
type, fall, 1970.

64.
65.
6 6

.

(67.

Student credit hours:
lecture
Student credit hours:
recitation
Student credit hours:
laboratory
Student credit hours:
Independent study or
credit classes)

Undergraduate
Undergraduate
Undergraduate
Undergraduate
variable

1Items in parentheses were eliminated from final
analysis.

130

Description

No.

.
69.
6 8

Student credit hours:
Graduate classes
Student credit hours:
Graduate ind.
study-var. credit classes
Teaching credits,
Fall, 1970.

70.
71.
72.
73.

Source

Teaching
Teaching
Teaching
Teaching

credits:
credits:
credits:
credits:

OIR Report:
Section Size
Analysis

total per section type;
'
Undergraduate lecture
Undergraduate recitation
Undergraduate laboratory
Graduate

Weighted average section size, per
section t y p e : F a l l , 1 9 7 0 . 1
74.
(75.
(76.
(77.

Weighted average section size:
Undergraduate lecture
Weighted average section size:
Undergraduate recitation)
Weighted average section size:
Undergraduate laboratory)
Weighted average section size:

Graduate)

Teaching load, per number of credits of
teaching and type of: i n s t r u c t i o n : F a l l ,
1970.

78.
(79.
80.
81.
82.
(83.
84.
85.
.

00
-J
•

8 6

•

00

00

Full-time
Full-time
Full-time
Full-time
Full-time
Full-time
Full-time
Full-time
credit:
Full-time
credit:
Full-time
credit:
Full-time
credit:

f a c u l t y , classes:
16.00+
13.00-15.99)
faculty, classes:
10.00-12.99
f a c u l t y , classes:
7.00-9.99
f a c u l t y , classes:
4.00-6.99
f a c u l t y , classes:
0.01-3.99)
f a c u l t y , classes:
0
f a c u l t y , classes:
f a c u l t y , ind. s t u d y - v a r .
16.00+
f a c u l t y , ind. s t u d y - v a r .
13.00-15.99
f a c u l t y , ind. study-var.
10.00-12.99
f a c u l t y , ind. study-var.
7.00-9. 99

OIR Report:
Teaching
Load & Time
Distribution
Analysis
II
II
II
II
II
It
It

II

II

II

II

C a l c u l a t e d by weighting the enrollment in each
section by the credit value of the section.

131

No.

89.
90.
91.
(92.
(93.
94.
95.
(96.
97.
98.
99.

100 .
101 .
102 .
103.
104.
105.
106.
107.
108.
109.

110

.

111

.

(112.

113.
114.
115.

Description

Source

Full-time faculty, ind. study-var.
credit:
4.00-6.99
Full-time faculty, ind. study-var.
0.01-3.99
credit:
Full-time faculty, ind. study-var.
credit:
0
Part-time faculty, classes:
16.00+)
Part-time faculty, classes:
13.00-15.99)
Part-time faculty, classes:
10.00-12.99
Part-time faculty, classes:
7.00-9.99
Part-time faculty, classes:
4.00-6.99)
Part-time faculty, classes:
0.01-3.99
Part-time faculty, classes:
0
Part-time faculty, ind. study-var.
credit:
16.00+
Part-time faculty, ind. study-var.
credit:
13.00-15.99
Part-time faculty, ind. study-var.
10.00-12.99
credit:
Part-time faculty, ind. study-var.
7.00-9.99
credit:
Part-time faculty, ind. study-var.
4.00-6.99
credit.
Part-time faculty, ind
study-var.
credit.
0.01-3.99
Part-time faculty, ind
study-var.
credit:
0
Graduate assistants, classes: 7.00-9.99
Graduate assistants, classes: 4.00-6.99
Graduate assistants, classes: 0.01-3.99
Graduate assistants, classes: 0
Graduate assistants, ind. study-var.
4.00-6.99
credit:
Graduate assistants, ind. study-var.
credit:
0.01-3.99
Graduate assistants, ind. study-var.
c r e d i t : 0)
Faculty body count:
fall, 1970.

Total academic staff,

Faculty body count:
Faculty body count:
Faculty body count:
assistants

full-time
part-time
graduate

O I R Report:
Teaching
Load & Time
Distribution
Analysis

132

No.

Description

Source

Full-time equivalent f a c u l t y ; Number
of full-time equivalent faculty
members paid from the general fund,
fall, 1970.
(116.
117.
118.

Full-time equivalent faculty:
Full-time equivalent faculty:
Full-time equivalent faculty:
assistants

Full-time)
Part-time
Graduate

Teaching credits, per type of staff
and type of in s t r u c t i o n : The total
credits of teaching for the academic
staff in a department,
fall, 1970.
119.

120 .
121 .
122 .
123.
124.

Teaching credits,
Classes
Teaching credits,
Ind. study-var.
Teaching credits,
Classes
Teaching credits,
Ind. study-var.
Teaching credits,
Classes
Teaching credits,
Ind. study-var.

full-time faculty:
full-time faculty:
credit
part-time faculty:
part-time faculty:
credit
graduate assistants;
graduate assistants
credit

Student credit hours, per type of staff
and type of i n s t r u c t i o n : Tne total
student credit hours taught by the
academic staff of a department, as of
the tenth day of classes, fall, 1970.
125.
126.
127.
128.
129.
130.

Student credit hours, full-time faculty:
Classes
Student credit hours, full-time faculty:
Ind. study-var. credit
Student credit hours, part-time faculty:
Classes
Student credit hours, part-time faculty:
Ind. study-var. credit
Student credit h o u r s , graduate
assistants:
Classes
Student credit h o u r s , graduate
assistants:
Ind. study-var. credit

O I R Report:
Teaching
Load & Time
Distribution
Analysis
II

133

No.

Source

Description
Time distribution, per FTE s t a f f ;
Percent
of staff time spent in
various activities, fall, 1970.

*131.
132.
133.
134.
*135.
136.
137.
(138.

Time distribution, FTE faculty:
[Not] Instruction
Time distribution, FTE faculty:
Research
Time distribution, FTE faculty:
Service
Time distribution, FTE faculty:
Administration
Time distribution, FTE graduate
assistants:
[Not] Instruction
Time distribution, FTE graduate
assistants:
Research
Time distribution, FTE graduate
assistants:
Service
Time distribution, FTE graduate
assistants:
Administration)
Student credit hours produced, per
course l e v e l : The total number of
student credit hours produced in a
department; term end credit hours
totals, including summer and offcampus courses, for 1969-70.

139.
140.
141.

Student credit hours produced,
u n d e r g r a d u a t e , lower division
Student credit hours produced,
undergraduate, upper division
Student credit hours produced, graduate
Faculty, full-time equivalent positions
(general f u n d T : For 1970-71.

142.
143.

Faculty, FTE positions
A-Faculty
Faculty, FTE positions
B-Faculty

(general fund):
(general fund):

*Reflected to negative form.

O I R Report:
Teaching
Load & Time
Distribution
Analysis
II
II
II

OIR:
Brown
Book, Fall,
1970, Tables
1.1-1.15

II
It
Brown Book,
Fall, 1970:
Tables 2.12.15

134

No.

Description
Catalog course
courses listed
1969-70.

144.
145.

Source

l i s t i n g s ; The number of
in the MSU annual catalog,

Catalog course
Catalog course

listings: Undergraduate
listings: Graduate

Courses t a u g h t : The number of courses
taught, total for 1969-70.

146.
147.
148.

Courses taught, undergraduate:
division
Courses taught, undergraduate:
division
Courses taught, graduate
General fund expenditures:
---------mn'fl.

149.

(152.
153.
(154.

Degrees g r a n t e d :

155.
156.
157.

158.
159.

Degrees granted:
Degrees granted:
Degrees granted:

If
Brown Book,
Fall, 1970:
Tables 5.15.15
II

Upper
II
II

Total for

General fund expenditures

Majors, total
Majors:
Percent
lower division
Majors:
Percent
upper division)
Majors:
Percent
Majors:
Percent

If

Lower

Departmental m a j o r s : Number of majors
as of seventh day of term, fall, 1969.

150.
151.

Brown Book,
Fall, 1970:
Tables 4.14.15

in undergraduate
in undergraduate
masters
doctors)
Total for 1969-70.

Bachelors
Masters
Doctors

Quality of graduate education:

1966

Quality of graduate education,
Faculty
Quality of graduate education,
Program

1966:

Brown Book,
Fall, 1970:
Tables 7.17.15
II
Brown B o o k ,
Fall, 1970:
Tables 9.19.15
II
II
II
II
II
Brown B o o k ,
Fall, 1970:
Tables 10.110.15
II
II

ACE:
"Cartter
Report"
II

1966:

135

No.

Source

Description
Quality of graduate e d u c a t i o n ;

160.
161.

Quality of graduate education,
Faculty
Quality of graduate education,
Program
Expenditures,

162.
*163.
164.
165.
166.
167.
168.
169.

1970.
1970:
1970:

171.

MSU Financial
Report

for 1969-70

Expenditures:
Total
[Non] Salaries:
Percent of total
expenditures (see var. 203)
Labor:
Percent of total expenditures
(see var. 204)
Supplies and services:
Percent of
total expenditures (see var. 205)
Equipment:
Percent of total expendi­
tures (see var. 206)
Experiment station funds:
Percent of
total expenditures (see var. 207)
Extension funds:
Percent of total
expenditures (see var. 208)
Educational Development Program f u n d s :
Percent of total expenditures
(see var. 209)
Sponsored research expenditures:
1969-70

170.

for

MSU Report:
Research Grant
and Contract
Expenditures
1969-70

Sponsored research f u n d s : Percent of
total expenditures (see var. 2 1 0 )
Sponsored research funds:
Total
Teaching certificate c a n d i d a t e s :
Percent of total u n d e r g r a d u a t e m a j o r s
seeking teaching certificates who are
in each department, for 1969-70.

172.

ACE:
"Roose
Re p o r t ”

Teaching certificate candidates:
Percent of university total
(see var. 2 1 1 )

♦Reflected to negative form.

Registrar,
Evaluation &
Research:
Teaching
Certificate
Candidates

136

No.

Description

Source

Student g e n d e r : Departmental majors
who are male, fall, 1970.

173.
174.

Male undergraduate majors:
Percent
of total
Male graduate majors:
Percent of
total
Faculty committee membership:
1970

(175 .

(178.

179.

*180.

(181.

Fall,

Fall,

1970

Faculty, total
Faculty ranks, professors:
Percent
of total faculty (see var. 2 1 2 )
Faculty ranks, associate professors:
Percent of total faculty (see
v a r . 213))
Faculty ranks, assistant professors:
Percent of total faculty (see
var. 214)
Faculty ranks, [not] instructors:
Percent of total faculty (see
var. 215)
Faculty ranks, others:
Percent of
total faculty (see var. 216)
Faculty degrees

182.

*183.

184.

MSU Directory

Faculty committee membership)
Faculty, size and r a n k s :

176.
177.

Registrar,
Evaluation
& Research:
Total Students

Faculty, highest degree:
Doctors,
percent of total faculty (see
var. 217)
Faculty, highest degree:
[Not]
masters, percent of total faculty
(see var. 218)
Faculty, highest degree:
Profes­
sional, percent of total faculty
(see var. 219)

*Reflected to negative form.

P r o v o s t 's
office

137

No.

Description
Faculty traits

(185.
186.
187.
188.
(189.

Faculty, age:
Born after 1926)
Faculty, gender:
Males, percent of
total
Faculty, marital status:
Married
Faculty, tenured
Faculty, degree from MSU)
Department majors, new and transfer:
Fall, 1970.

190.
191.
192.
(193.
(194.

204.
205.
206.
207.
208.
209.

(210.

II
II
II
II
Registrar:
New students

OIR Report:
Clerical &
Technical Aid

Clerical aides, FTE:
General fund)
Other funds
Clerical aides, FTE:
General fund)
Technical aides, FTE:
Technical aides, FTE:
Other funds
E x p e n d i t u r e s , for 1969-70

203.

Provost's
office

Majors, new:
Percent of total majors
Majors, new undergraduates:
Percent
of total undergraduate majors
Majors, transfer undergraduates:
Percent of total undergraduates
Majors, new masters:
Percent of total
masters)
Majors, new doctors:
Percent of total
doctors)
Aides, clerical and technical, iper
source of s u p p o r t : Number of aides
assigned, August, 1971.

(195.
196.
(197.
198.

Source

Salaries, total expenditures
(see var. 163)
Labor, total expenditures
(see var. 164)
Supplies and services, total
expenditures (see var. 165)
Equipment, total expenditures
(see var. 166)
Experiment station funds, total
(see var. 167)
Extension funds, total (see var. 168)
Educational Development Program funds
(see var. 169)
Sponsored research funds (see var. 170))

MSU Financial
Report

138

Description

Source

Teaching certificate candidates:

Registrar,
Evaluation
& Research:
Teaching
Certificate
Candidates

No.

1^ 9-70

211 . Teaching certificate candidates,
total
Faculty,

212 .
213.
214.
*215.
(216.

(see var.

172)

size and r a n k s :

Fall,

Faculty ranks, professor:
Total
(see var. 177)
Faculty ranks, associate professor:
Total (see var. 178)
Faculty ranks, assistant professor:
Total (see var. 179)
Faculty ranks, (not] instructor:
Total (see var. 180)
Faculty ranks, other:
Total
(see var. 181))
F aculty degrees

217.
*218.
219.

Faculty,
total
Faculty,
total
Faculty,
total

highest degree:
(see var. 182)
highest degree:
(see var. 183)
highest degree:
(see var. 184)

Doctors,
[Not] masters,
Professional,

*Reflected to negative form.

II
P r o v o s t 's
office

APPENDIX C

PROGRAMS AND PROCEDURES THAT MADE
UP THE METHODOLOGY

APPENDIX C

PROGRAMS AND PROCEDURES THAT MADE
UP THE METHODOLOGY

1.

The following routines were part of the PACKAGE

program used in the present study:

(1) CORR:

A program which computes means, standard
deviations, and product-moment correlations
for up to 150 variables.

(2) ORDER:

A program w hich arrays the variables in the
correlation matrix stored in the computer's
core memory in a new order starting with
the highest correlation and linking each
variable sequentially with the one with
which it most highly correlates.

(3) SQRR:

A program which produces a matrix of sim­
ilarity coefficients, i.e., interrelates
the variables in terms of the similarities
in their patterns of correlations.

(4) ARRANGE:

A program that takes a specified matrix
stored on tape (e.g., correlation) and puts
that matrix in the order of a matrix that
is in the computer core (e.g., SQRR).

(5) REORDER:

A program that puts a matrix in any order
which is specified; which deletes variables
from a matrix.

(6 ) MGRP:

A program which performs an oblique
multiple-groups analysis on groups speci­
fied from the variables comprising the
correlational matrix in the computer's
core memory; communalities are used in
the diagonal of the matrix.

139

140

2.

The following steps in the procedures formed the

method used to amalgamate the departmental traits and the
de p a r t m e n t s :

Clustering Variables
Step 1

Run
Run
Run

1

Run
Run

4

Step 3

Run

6

Addition of variables 199-219

Step 4

Run

7

Combination of variables 1-77 and 78-219
into final two sets of clusters

Step 2

2
3

Clustering variables 1-77
Clustering
variables 1-77
Clustering
variables 1-77

5

Clustering variables 78-198
Clustering
variables 78-198

8
9

10
11
12
13
14
15

Grouping Departments
Step 5

Run 16

Grouping departments into four sets
of groups

17
18
Step

6

Run 19
20
22

Grouping of departments in Set A
averaging)

(percents,

APPENDIX D

CLUSTERS OF DEPARTMENTAL TRAITS

APPENDIX D

CLUSTERS OF DEPARTMENTAL TRAITS

CLUSTER 1
MASTERS
*5 NUM OF COURSES
DOCTORS
* 6 NUM OF COURSES
*9 COURSE ENROLLMENTS— UPPER DIVISION
*11 COURSE ENROLLMENTS— MASTERS
*12 COURSE ENROLLMENTS— DOCTORS
15 CLASS IJOURS--- UPPER DIVISION
17 CLASS H OURS--- MASTERS
18 CLASS HOURS--- DOCTORS
IN ORG. CLASSES-- UPPER DIVISION
*21 TEACHING CREDITS
IN ORG. CLASSES-- MASTERS
*23 TEACHING CREDITS
IN ORG. CLASSES-- DOCTORS
*24 TEACHING CREDITS
27 TEACHING CREDITS IN IND-VAR-- UPPER DIVISION
28 TEACHING CREDITS IN IND-VAR-- MASTERS
*29 TEACHING CREDITS IN IND-VAR-- DOCTORS
*32 STUDENT CREDIT HOURS-UPPER DIVISION
*34 STUDENT CREDIT HOURS-MASTERS
**35 STUDENT CREDIT HOURS-DOCTORS
36 SECTION SIZE— LECTURE--- 201+
43 SECTION SIZE— RECITATION--- 201+
62 SECTION SIZE— GRADUATE— 11-20
64 TOTAL STUDENT CREDIT H OURS— UNDERGRAD--- LECTURE
68 TOTAL STUDENT CREDIT HOURS— GRADUATE-- CLASSES
*69 TOTAL STUDENT CREDIT HOURS— GRADUATE-- IND-VAR
*73 TOTAL CREDITS OF TEACHING— GRADUATE--74 WEIGHTED AVERAGE SIZE OF SECTIONS--- UNDERGRAD— LECTURE
84 NUM FULL-TIME FACULTY WITH CLASS LOAD 0
85 NUM FULL-TIME FACULTY WITH IND-VAR LOAD 16.00+
86 NUM FULL-TIME FACULTY WITH IND-VAR LOAD 13.00-15.99
*87 NUM FULL-TIME FACULTY WITH IND-VAR LOAD 10.00-12.99
FACULTY WITH IND-VAR LOAD 7.00- 9.99
*88 NUM FULL-TIME
99 NUM PART-TIME FACULTY WITH IND-VAR LOAD 16.00+

*Loading on trait-cluster of 67-85.
**Loading on trait-cluster of

141

86

or higher.

142

107
108
*115
**118
*120
*123
126
*129
*140
*141
*143
*145
*148
*149
*150
155
156
*157
158
159
160
161

NUM GRAD ASSISTANTS WITH CLASS LOAD
4.00-6.99
NUM GRAD ASSISTANTS WITH CLASS LOAD
0.01-3.99
NUM GRAD ASSISTANTS
NUM FULL-TIME EQUIV FACULTY OCCUPIED BY GRAD ASSISTANTS
TOTAL TEACHING CREDITS FULL-TIME FACULTY
IND-VAR
TOTAL TEACHING CREDITS GRAD ASSISTANTS
CLASSES
TOTAL STUDENT CREDIT HOURS FULL-TIME FACULTY
IND-VAR
TOTAL STUDENT CREDIT HOURS GRAD ASSISTANTS
CLASSES
STUDENT CREDIT HOURS PRODUCED BY DEPT--- UPPER DIVISION
STUDENT CREDIT HOURS PRODUCED BY DEPT--- GRADUATE
NUM FTE FACULTY(GENERAL FUND)
B-FACULTY
NUM CATALOG COURSE LISTINGS
GRAD
NUM COURSES TAUGHT BY DEPT
GRADUATE
GENERAL FUND EXPENDITURES BY DEPT(MULTIPLY BY 100)
TOTAL NUM DEPARTMENT MAJORS
NUM DEGREES GRANTED BY DEPT--BACHELOR
NUM DEGREES GRANTED BY DEPT--MASTERS
NUM DEGREES GRANTED BY DEPT--DOCTORS
QUALITY OF GRADUATE EDUCATION(CARRTER)--- FACULTY
QUALITY OF GRADUATE EDUCATION(CARRTER)--- PROGRAM
QUALITY OF GRADUATE EDUCATION(ROOSE)--- FACULTY
QUALITY OF GRADUATE EDUCATION(ROOSE)--- PROGRAM

CLUSTER 2
*2
**3
*48
*49
63
106
**144
146
*147
*172

NUM OF COURSES
LOWER DIVISION
NUM OF COURSES
UPPER DIVISION
SECTION SIZE— RECITATION
11-20
SECTION SIZE— RECITATION
1-10
SECTION SIZE— GRADUATE— 1-10
NUM GRAD ASSISTANTS WITH CLASS LOAD
7.00-9.99
NUM CATALOG COURSE LISTINGS
UNDERGRAD
NUM COURSES TAUGHT BY DEPT
LOWER DIVISION
NUM COURSES TAUGHT BY DEPT
UPPER DIVISION
PER CENT UNDERGRAD MAJORS SEEKING TEACHING CERTIFICATES

CLUSTER 3
**8

*14
**20
**31
45
46
*47
**65
**71
78
80

COURSE ENROLLMENTS— LOWER DIVISION
CLASS H OURS
LOWER DIVISION
TEACHING CREDITS IN ORG. CLASSES
LOWER DIVISION
STUDENT CREDIT HOURS
LOWER DIVISION
SECTION SIZE— RECITATION-- 51-100
SECTION SIZE— RECITATION-- 36-50
SECTION SIZE— RECITATION-- 21-35
TOTAL STUDENT CREDIT HOURS— UNDERGRAD
RECITATION
TOTAL CREDITS OF TEACHING— UNDERGRAD
RECITATION
NUM FULL-TIME FACULTY WITH CLASS LOAD 16.00+
NUM FULL-TIME FACULTY WITH CLASS LOAD 10.00-12.99

143

81
82
89
*90
*91
94
*95
**113
**119
*121
**125
**139
**142
176

NUM FULL-TIME FACULTY WITH CLASS LOAD
7.00- 9.99
NUM FULL-TIME FACULTY WIT H CLASS LOAD
4.00-6.99
NUM FULL-TIME FACULTY WITH IND-VAR LOAD 4.00-6.99
NUM FULL-TIME FACULTY WITH IND-VAR LOAD 0.01-3.99
NUM FULL-TIME FACULTY WITH IND-VAR LOAD 0
NUM PART-TIME FACULTY WITH CLASS LOAD 10.00-12.99
NUM PART-TIME FACULTY WITH CLASS LOAD
7.00-9.99
NUM FULL-TIME FACULTY
TOTAL TEACHING CREDITS FULL-TIME FACULTY
CLASSES
TOTAL TEACHING CREDITS PART-TIME FACULTY
CLASSES
TOTAL STUDENT CREDIT HOURS FULL-TIME FACULTY
CLASSES
STUDENT CREDIT HOURS PRODUCED BY DEPT
LOWER DIVISION
NUM FTE FACULTY(GENERAL FUND)---- A-FACULTY
TOTAL FACULTY

CLUSTER 4
0 . 0 1 -3.99
97 NUM PART-TIME FACULTY WITH CLASS LOAD
0•
**98 NUM PART-TIME FACULTY WITH CLASS LOAD
1 0 0 NUM PART-TIME FACULTY WITH IND-VAR LOAD 13. 00-15 .99
1 0 1 NUM PART-TIME FACULTY WITH IND-VAR LOAD 1 0 .0 0 - 1 2 .99
7. 00-9. 99
1 0 2 NUM PART-TIME FACULTY WITH IND-VAR LOAD
4. 0 0 - 6 .99
*103 NUM PART-TIME FACULTY WITH IND-VAR LOAD
0 .01-3. 99
104 NUM PART-TIME FACULTY WITH IND-VAR LOAD
0
*105 NUM PART-TIME FACULTY WITH IND-VAR LOAD
*109 NUM GRAD ASSISTANTS WITH CLASS LOAD
0
**114 NUM PART-TIME FACULTY
*117 NUM FULL-TIME EQUIV FACULTY OCCUPIED BY PART-TIME FACULTY
**122 TOTAL TEACHING CREDITS PART-TIME FACULTY
IND-VAR
127 TOTAL STUDENT CREDIT HOURS PART-TIME FACULTY--- -CLASSES
*128 TOTAL STUDENT CREDIT HOURS PART-TIME FACULTY---- IND-VAR
*162 TOTAL EXPENDITURES(MULTIPLY BY 100)
171 SPONSORED RESEARCH EXPENDITURES
196 FTE CLERICAL AIDS, OTHER FUND
198 FTE TECHNICAL AIDS, OTHER FUND

CLUSTER 5
164 LABOR AS PER CENT OF TOTAL EXPENDITURE
**167 EXPERIMENT STATION FUNDS AS PER CENT OF TOTAL EXPENDITURE
16 8 EXTENSION FUNDS AS PER CENT OF TOTAL EXPENDITURE

144

CLUSTER 6
**163
**165
*166
**170

NON-SALARY EXPENDITURES AS PER CENT OF TOTAL EXPENDITURES
SUPPLIES AND SERVICE AS PER CENT OF TOTAL EXPENDITURE
EQUIPMENT AS PER CENT OF TOTAL EXPENDITURES
SPONSORED RESEARCH FUNDS AS P E R CENT OF TOTAL EXPENDITURE

CLUSTER 7
**131
**132
135
*136

DIST
DIST
DIST
DIST

TIME
TIME
TIME
TIME

CLUSTER

(PER
(PER
(PER
(PER

CENT)
CENT)
CENT)
CENT)

FTE
FTE
FTE
FTE

FACULTY— NON-INSTRUCTION
FACULTY
RESEARCH
'
GRAD ASSIST--- NON-INSTRUCTION
GRAD ASSIST--- RESEARCH

8

*134 TIME DIST (PER CENT) FTE FACULTY
ADMINISTRATION
179 FACULTY RANKS
PER CENT ASSISTANTS
*184 F A C U L T Y — HIGHEST DEGREE
PER CENT PROFESSIONAL

CLUSTER 9
**26
*51
*52
*110
**111
**124
130

TEACHING CREDITS IN IND-VAR
LOWER DIVISION
SECTION SIZE— LABORATORY
101-200
SECTION SIZE— LABORATORY
51-100
NUM GRAD ASSISTANTS WITH IND-VAR LOAD 4.00-6.99
NUM GRAD ASSISTANTS WITH IND-VAR LOAD 0.01-3.99
TOTAL TEACHING CREDITS GRAD ASSISTANTSIND-VAR
TOTAL STUDENT CREDIT HOURS GRAD A SSISTANTS
IND-VAR

CLUSTER 1C
*37
*38
39
*53
*54
55
56

SECTION
SIZE— LECTURE---101-200
SECTION
SIZE— LECTURE ---51-100
SECTION
SIZE — LECTURE---36-50
SECTION
SIZE— LABORATORY----- 36-50
SECTION
SIZE— LABORATORY----- 21-35
SECTION
SIZE— LABORATORY ----- 11-20
SECTION
SIZE— LABORATORY----- 1-10
**66
TOTAL STUDENT CREDIT H OURS— UNDERGRAD
LABORATORY
*70 TOTAL CREDITS OF TEACHING— UNDERGRAD— LECTURE
**72 TOTAL CREDITS OF TEACHING— UNDERGRAD
LABORATORY

145

CLUSTER 11
*173
*174
*177
*180
*182
183
*186
187
*188

PER CENT UNDERGRAD MALES
PER CENT GRAD MALES
FACULTY RANKS-- PER CENT PROFESSORS
FACULTY RANKS-- PER CENT NON-INSTRUCTORS
FACULTY— HIGHEST DEGREE
PER CENT DOCTORS
FACULTY— HIGHEST DEGREE-- PER CENT MASTERS
FACULTY SEX (PERCENT MALE)
FACULTY— PERCENT MARRIED
FACULTY— PERCENT TENURED

CLUSTER 12
*191
151
*190
192

NEW
PER
NEW
NEW

AND TRANSFER MAJORS
PERCENT NEW UNDERGRADS
CENT OF MAJORS IN LOWER DIVISION
AND TRANSFER MAJORS
PERCENT NEW MAJORS
AND TRANSFER MAJORS
PERCENT UNDERGRAD TRANSFERS

CLUSTER 13
*1
**7
**13
**19
25
**30
40
41

NUM OF COURSES
SUB-COLLEGE
COURSE ENROLLMENTS
SUB-COLLEGE
CLASS HOURS
SUB-COLLEGE
TEACHING CREDITS IN ORG. CLASSES
SUB-COLLEGE
TEACHING CREDITS IN IND-VAR
SUB-COLLEGE
STUDENT CREDIT HOURS
SUB-COLLEGE
SECTION SIZE— LECTURE
21-35
SECTION SIZE— LECTURE
11-20

CLUSTER 14
**4
**10
**16
**22
**33
*153

NU M OF COURSES
GRAD-PROF
COURSE ENROLLMENTS— GRAD-PROF
CLASS HOURS
GRAD-PROF
TEACHING CREDITS IN ORG. CLASSES
STUDENT CREDIT HOURS— GRAD-PROF
PER CENT OF MAJORS IN MASTERS

GRAD-PROF

CLUSTER 15
169 EDUCATIONAL DEVELOPMENT PROGRAM AS PER CENT OF TOTAL
EXPENDITURE
CLUSTER 16
*133 TIME DIST(PER CENT)
*137 TIME DIST(PER CENT)

FTE FACULTY
SERVICE
FTE GRAD ASSIST
SERVICE